From 7723d291ce59891d7db4f2eb5dcf4a0884bd83ac Mon Sep 17 00:00:00 2001 From: "renovate[bot]" <29139614+renovate[bot]@users.noreply.github.com> Date: Sun, 1 Mar 2026 01:14:16 +0000 Subject: [PATCH 01/38] chore(deps): update dependency @types/node to ^25.3.3 --- frontend/package-lock.json | 8 ++++---- frontend/package.json | 2 +- package-lock.json | 8 ++++---- package.json | 2 +- 4 files changed, 10 insertions(+), 10 deletions(-) diff --git a/frontend/package-lock.json b/frontend/package-lock.json index 505b725f..f1ab2cb8 100644 --- a/frontend/package-lock.json +++ b/frontend/package-lock.json @@ -41,7 +41,7 @@ "@testing-library/jest-dom": "^6.9.1", "@testing-library/react": "^16.3.2", "@testing-library/user-event": "^14.6.1", - "@types/node": "^25.3.2", + "@types/node": "^25.3.3", "@types/react": "^19.2.14", "@types/react-dom": "^19.2.3", "@typescript-eslint/eslint-plugin": "^8.56.1", @@ -3565,9 +3565,9 @@ "license": "MIT" }, "node_modules/@types/node": { - "version": "25.3.2", - "resolved": "https://registry.npmjs.org/@types/node/-/node-25.3.2.tgz", - "integrity": "sha512-RpV6r/ij22zRRdyBPcxDeKAzH43phWVKEjL2iksqo1Vz3CuBUrgmPpPhALKiRfU7OMCmeeO9vECBMsV0hMTG8Q==", + "version": "25.3.3", + "resolved": "https://registry.npmjs.org/@types/node/-/node-25.3.3.tgz", + "integrity": "sha512-DpzbrH7wIcBaJibpKo9nnSQL0MTRdnWttGyE5haGwK86xgMOkFLp7vEyfQPGLOJh5wNYiJ3V9PmUMDhV9u8kkQ==", "dev": true, "license": "MIT", "dependencies": { diff --git a/frontend/package.json b/frontend/package.json index ccafb968..79ec151e 100644 --- a/frontend/package.json +++ b/frontend/package.json @@ -60,7 +60,7 @@ "@testing-library/jest-dom": "^6.9.1", "@testing-library/react": "^16.3.2", "@testing-library/user-event": "^14.6.1", - "@types/node": "^25.3.2", + "@types/node": "^25.3.3", "@types/react": "^19.2.14", "@types/react-dom": "^19.2.3", "@typescript-eslint/eslint-plugin": "^8.56.1", diff --git a/package-lock.json b/package-lock.json index 23f89488..3143f390 100644 --- a/package-lock.json +++ b/package-lock.json @@ -14,7 +14,7 @@ "devDependencies": { "@bgotink/playwright-coverage": "^0.3.2", "@playwright/test": "^1.58.2", - "@types/node": "^25.3.2", + "@types/node": "^25.3.3", "dotenv": "^17.3.1", "markdownlint-cli2": "^0.21.0", "prettier": "^3.8.1", @@ -937,9 +937,9 @@ "license": "MIT" }, "node_modules/@types/node": { - "version": "25.3.2", - "resolved": "https://registry.npmjs.org/@types/node/-/node-25.3.2.tgz", - "integrity": "sha512-RpV6r/ij22zRRdyBPcxDeKAzH43phWVKEjL2iksqo1Vz3CuBUrgmPpPhALKiRfU7OMCmeeO9vECBMsV0hMTG8Q==", + "version": "25.3.3", + "resolved": "https://registry.npmjs.org/@types/node/-/node-25.3.3.tgz", + "integrity": "sha512-DpzbrH7wIcBaJibpKo9nnSQL0MTRdnWttGyE5haGwK86xgMOkFLp7vEyfQPGLOJh5wNYiJ3V9PmUMDhV9u8kkQ==", "devOptional": true, "license": "MIT", "dependencies": { diff --git a/package.json b/package.json index 7c640572..b46bfeb2 100644 --- a/package.json +++ b/package.json @@ -19,7 +19,7 @@ "devDependencies": { "@bgotink/playwright-coverage": "^0.3.2", "@playwright/test": "^1.58.2", - "@types/node": "^25.3.2", + "@types/node": "^25.3.3", "dotenv": "^17.3.1", "markdownlint-cli2": "^0.21.0", "prettier": "^3.8.1", From a83967daa33af2b0f8dfa6fa9bc73c1512511ed0 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 01:15:54 +0000 Subject: [PATCH 02/38] fix(deps): add new dependencies for pbkdf2, scram, stringprep, and pkcs8 --- go.work.sum | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/go.work.sum b/go.work.sum index 7e4b3b20..ebcc3c1e 100644 --- a/go.work.sum +++ b/go.work.sum @@ -70,8 +70,12 @@ github.com/spf13/viper v1.15.0/go.mod h1:fFcTBJxvhhzSJiZy8n+PeW6t8l+KeT/uTARa0jH github.com/stretchr/testify v1.4.0/go.mod h1:j7eGeouHqKxXV5pUuKE4zz7dFj8WfuZ+81PSLYec5m4= github.com/subosito/gotenv v1.4.2 h1:X1TuBLAMDFbaTAChgCBLu3DU3UPyELpnF2jjJ2cz/S8= github.com/subosito/gotenv v1.4.2/go.mod h1:ayKnFf/c6rvx/2iiLrJUk1e6plDbT3edrFNGqEflhK0= +github.com/xdg-go/pbkdf2 v1.0.0/go.mod h1:jrpuAogTd400dnrH08LKmI/xc1MbPOebTwRqcT5RDeI= +github.com/xdg-go/scram v1.2.0/go.mod h1:3dlrS0iBaWKYVt2ZfA4cj48umJZ+cAEbR6/SjLA88I8= +github.com/xdg-go/stringprep v1.0.4/go.mod h1:mPGuuIYwz7CmR2bT9j4GbQqutWS1zV24gijq1dTyGkM= github.com/xhit/go-str2duration/v2 v2.1.0 h1:lxklc02Drh6ynqX+DdPyp5pCKLUQpRT8bp8Ydu2Bstc= github.com/xhit/go-str2duration/v2 v2.1.0/go.mod h1:ohY8p+0f07DiV6Em5LKB0s2YpLtXVyJfNt1+BlmyAsU= +github.com/youmark/pkcs8 v0.0.0-20240726163527-a2c0da244d78/go.mod h1:aL8wCCfTfSfmXjznFBSZNN13rSJjlIOI1fUNAtF7rmI= github.com/yuin/goldmark v1.4.13 h1:fVcFKWvrslecOb/tg+Cc05dkeYx540o0FuFt3nUVDoE= github.com/yuin/goldmark v1.4.13/go.mod h1:6yULJ656Px+3vBD8DxQVa3kxgyrAnzto9xy5taEt/CY= golang.org/x/crypto v0.44.0/go.mod h1:013i+Nw79BMiQiMsOPcVCB5ZIJbYkerPrGnOa00tvmc= From 1a559e3c64049b1e885b44698ba0ec80743dc4db Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 01:31:48 +0000 Subject: [PATCH 03/38] fix(deps): update caniuse-lite to version 1.0.30001775 --- frontend/package-lock.json | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/frontend/package-lock.json b/frontend/package-lock.json index f1ab2cb8..b8d9823b 100644 --- a/frontend/package-lock.json +++ b/frontend/package-lock.json @@ -4350,9 +4350,9 @@ } }, "node_modules/caniuse-lite": { - "version": "1.0.30001774", - "resolved": "https://registry.npmjs.org/caniuse-lite/-/caniuse-lite-1.0.30001774.tgz", - "integrity": "sha512-DDdwPGz99nmIEv216hKSgLD+D4ikHQHjBC/seF98N9CPqRX4M5mSxT9eTV6oyisnJcuzxtZy4n17yKKQYmYQOA==", + "version": "1.0.30001775", + "resolved": "https://registry.npmjs.org/caniuse-lite/-/caniuse-lite-1.0.30001775.tgz", + "integrity": "sha512-s3Qv7Lht9zbVKE9XoTyRG6wVDCKdtOFIjBGg3+Yhn6JaytuNKPIjBMTMIY1AnOH3seL5mvF+x33oGAyK3hVt3A==", "dev": true, "funding": [ { From e90ad34c289c9bf7da123d9b6d895d9289111d1a Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 01:33:26 +0000 Subject: [PATCH 04/38] chore: add script to update Go module dependencies --- scripts/go_update.sh | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) create mode 100755 scripts/go_update.sh diff --git a/scripts/go_update.sh b/scripts/go_update.sh new file mode 100755 index 00000000..8df50953 --- /dev/null +++ b/scripts/go_update.sh @@ -0,0 +1,16 @@ +#!/bin/bash + +# This script updates Go module dependencies for the project. + +cd /projects/Charon/backend || exit + +echo "Updating Go module dependencies..." + +go get -u ./... +go mod tidy +go mod verify +go vet ./... +go list -m -u all +go build ./... + +echo "Go module dependencies updated successfully." From b78798b877a83503d6020c5a577a87081493eeca Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 01:34:37 +0000 Subject: [PATCH 05/38] chore: Update dependencies in go.sum - Bump github.com/bytedance/sonic from v1.14.1 to v1.15.0 - Bump github.com/gabriel-vasile/mimetype from v1.4.12 to v1.4.13 - Bump github.com/glebarez/go-sqlite from v1.21.2 to v1.22.0 - Bump github.com/gin-gonic/gin from v1.11.0 to v1.12.0 - Bump github.com/google/pprof to v0.0.0-20250317173921-a4b03ec1a45e - Bump go.opentelemetry.io/auto/sdk to v1.2.1 - Bump go.opentelemetry.io/otel to v1.40.0 - Update various other dependencies to their latest versions --- backend/go.mod | 31 ++++++------- backend/go.sum | 120 ++++++++++++++++++++++++++++--------------------- 2 files changed, 84 insertions(+), 67 deletions(-) diff --git a/backend/go.mod b/backend/go.mod index b6b8267c..75ec8a47 100644 --- a/backend/go.mod +++ b/backend/go.mod @@ -42,9 +42,9 @@ require ( github.com/docker/go-units v0.5.0 // indirect github.com/dustin/go-humanize v1.0.1 // indirect github.com/felixge/httpsnoop v1.0.4 // indirect - github.com/gabriel-vasile/mimetype v1.4.12 // indirect + github.com/gabriel-vasile/mimetype v1.4.13 // indirect github.com/gin-contrib/sse v1.1.0 // indirect - github.com/glebarez/go-sqlite v1.21.2 // indirect + github.com/glebarez/go-sqlite v1.22.0 // indirect github.com/go-logr/logr v1.4.3 // indirect github.com/go-logr/stdr v1.2.2 // indirect github.com/go-playground/locales v0.14.1 // indirect @@ -66,6 +66,7 @@ require ( github.com/modern-go/reflect2 v1.0.2 // indirect github.com/morikuni/aec v1.0.0 // indirect github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 // indirect + github.com/ncruces/go-strftime v1.0.0 // indirect github.com/opencontainers/go-digest v1.0.0 // indirect github.com/opencontainers/image-spec v1.1.1 // indirect github.com/oschwald/maxminddb-golang/v2 v2.1.1 // indirect @@ -73,8 +74,8 @@ require ( github.com/pkg/errors v0.9.1 // indirect github.com/pmezard/go-difflib v1.0.0 // indirect github.com/prometheus/client_model v0.6.2 // indirect - github.com/prometheus/common v0.66.1 // indirect - github.com/prometheus/procfs v0.16.1 // indirect + github.com/prometheus/common v0.67.5 // indirect + github.com/prometheus/procfs v0.20.1 // indirect github.com/quic-go/qpack v0.6.0 // indirect github.com/quic-go/quic-go v0.59.0 // indirect github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec // indirect @@ -82,20 +83,20 @@ require ( github.com/twitchyliquid64/golang-asm v0.15.1 // indirect github.com/ugorji/go/codec v1.3.1 // indirect go.mongodb.org/mongo-driver/v2 v2.5.0 // indirect - go.opentelemetry.io/auto/sdk v1.1.0 // indirect - go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.63.0 // indirect - go.opentelemetry.io/otel v1.38.0 // indirect + go.opentelemetry.io/auto/sdk v1.2.1 // indirect + go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.65.0 // indirect + go.opentelemetry.io/otel v1.40.0 // indirect go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.38.0 // indirect - go.opentelemetry.io/otel/metric v1.38.0 // indirect - go.opentelemetry.io/otel/trace v1.38.0 // indirect - go.yaml.in/yaml/v2 v2.4.2 // indirect - golang.org/x/arch v0.22.0 // indirect + go.opentelemetry.io/otel/metric v1.40.0 // indirect + go.opentelemetry.io/otel/trace v1.40.0 // indirect + go.yaml.in/yaml/v2 v2.4.3 // indirect + golang.org/x/arch v0.24.0 // indirect golang.org/x/sys v0.41.0 // indirect google.golang.org/protobuf v1.36.11 // indirect gopkg.in/yaml.v3 v3.0.1 // indirect gotest.tools/v3 v3.5.2 // indirect - modernc.org/libc v1.22.5 // indirect - modernc.org/mathutil v1.5.0 // indirect - modernc.org/memory v1.5.0 // indirect - modernc.org/sqlite v1.23.1 // indirect + modernc.org/libc v1.68.1 // indirect + modernc.org/mathutil v1.7.1 // indirect + modernc.org/memory v1.11.0 // indirect + modernc.org/sqlite v1.46.1 // indirect ) diff --git a/backend/go.sum b/backend/go.sum index db8c59b6..1fed2afc 100644 --- a/backend/go.sum +++ b/backend/go.sum @@ -6,12 +6,8 @@ github.com/beorn7/perks v1.0.1 h1:VlbKKnNfV8bJzeqoa4cOKqO6bYr3WgKZxO8Z16+hsOM= github.com/beorn7/perks v1.0.1/go.mod h1:G2ZrVWU2WbWT9wwq4/hrbKbnv/1ERSJQ0ibhJ6rlkpw= github.com/bytedance/gopkg v0.1.3 h1:TPBSwH8RsouGCBcMBktLt1AymVo2TVsBVCY4b6TnZ/M= github.com/bytedance/gopkg v0.1.3/go.mod h1:576VvJ+eJgyCzdjS+c4+77QF3p7ubbtiKARP3TxducM= -github.com/bytedance/sonic v1.14.1 h1:FBMC0zVz5XUmE4z9wF4Jey0An5FueFvOsTKKKtwIl7w= -github.com/bytedance/sonic v1.14.1/go.mod h1:gi6uhQLMbTdeP0muCnrjHLeCUPyb70ujhnNlhOylAFc= github.com/bytedance/sonic v1.15.0 h1:/PXeWFaR5ElNcVE84U0dOHjiMHQOwNIx3K4ymzh/uSE= github.com/bytedance/sonic v1.15.0/go.mod h1:tFkWrPz0/CUCLEF4ri4UkHekCIcdnkqXw9VduqpJh0k= -github.com/bytedance/sonic/loader v0.3.0 h1:dskwH8edlzNMctoruo8FPTJDF3vLtDT0sXZwvZJyqeA= -github.com/bytedance/sonic/loader v0.3.0/go.mod h1:N8A3vUdtUebEY2/VQC0MyhYeKUFosQU6FxH2JmUe6VI= github.com/bytedance/sonic/loader v0.5.0 h1:gXH3KVnatgY7loH5/TkeVyXPfESoqSBSBEiDd5VjlgE= github.com/bytedance/sonic/loader v0.5.0/go.mod h1:AR4NYCk5DdzZizZ5djGqQ92eEhCCcdf5x77udYiSJRo= github.com/cenkalti/backoff/v5 v5.0.3 h1:ZN+IMa753KfX5hd8vVaMixjnqRZ3y8CuJKRKj1xcsSM= @@ -41,18 +37,16 @@ github.com/dustin/go-humanize v1.0.1 h1:GzkhY7T5VNhEkwH0PVJgjz+fX1rhBrR7pRT3mDkp github.com/dustin/go-humanize v1.0.1/go.mod h1:Mu1zIs6XwVuF/gI1OepvI0qD18qycQx+mFykh5fBlto= github.com/felixge/httpsnoop v1.0.4 h1:NFTV2Zj1bL4mc9sqWACXbQFVBBg2W3GPvqp8/ESS2Wg= github.com/felixge/httpsnoop v1.0.4/go.mod h1:m8KPJKqk1gH5J9DgRY2ASl2lWCfGKXixSwevea8zH2U= -github.com/gabriel-vasile/mimetype v1.4.12 h1:e9hWvmLYvtp846tLHam2o++qitpguFiYCKbn0w9jyqw= -github.com/gabriel-vasile/mimetype v1.4.12/go.mod h1:d+9Oxyo1wTzWdyVUPMmXFvp4F9tea18J8ufA774AB3s= +github.com/gabriel-vasile/mimetype v1.4.13 h1:46nXokslUBsAJE/wMsp5gtO500a4F3Nkz9Ufpk2AcUM= +github.com/gabriel-vasile/mimetype v1.4.13/go.mod h1:d+9Oxyo1wTzWdyVUPMmXFvp4F9tea18J8ufA774AB3s= github.com/gin-contrib/gzip v1.2.5 h1:fIZs0S+l17pIu1P5XRJOo/YNqfIuPCrZZ3TWB7pjckI= github.com/gin-contrib/gzip v1.2.5/go.mod h1:aomRgR7ftdZV3uWY0gW/m8rChfxau0n8YVvwlOHONzw= github.com/gin-contrib/sse v1.1.0 h1:n0w2GMuUpWDVp7qSpvze6fAu9iRxJY4Hmj6AmBOU05w= github.com/gin-contrib/sse v1.1.0/go.mod h1:hxRZ5gVpWMT7Z0B0gSNYqqsSCNIJMjzvm6fqCz9vjwM= -github.com/gin-gonic/gin v1.11.0 h1:OW/6PLjyusp2PPXtyxKHU0RbX6I/l28FTdDlae5ueWk= -github.com/gin-gonic/gin v1.11.0/go.mod h1:+iq/FyxlGzII0KHiBGjuNn4UNENUlKbGlNmc+W50Dls= github.com/gin-gonic/gin v1.12.0 h1:b3YAbrZtnf8N//yjKeU2+MQsh2mY5htkZidOM7O0wG8= github.com/gin-gonic/gin v1.12.0/go.mod h1:VxccKfsSllpKshkBWgVgRniFFAzFb9csfngsqANjnLc= -github.com/glebarez/go-sqlite v1.21.2 h1:3a6LFC4sKahUunAmynQKLZceZCOzUthkRkEAl9gAXWo= -github.com/glebarez/go-sqlite v1.21.2/go.mod h1:sfxdZyhQjTM2Wry3gVYWaW072Ri1WMdWJi0k6+3382k= +github.com/glebarez/go-sqlite v1.22.0 h1:uAcMJhaA6r3LHMTFgP0SifzgXg46yJkgxqyuyec+ruQ= +github.com/glebarez/go-sqlite v1.22.0/go.mod h1:PlBIdHe0+aUEFn+r2/uthrWq4FxbzugL0L8Li6yQJbc= github.com/glebarez/sqlite v1.11.0 h1:wSG0irqzP6VurnMEpFGer5Li19RpIRi2qvQz++w0GMw= github.com/glebarez/sqlite v1.11.0/go.mod h1:h8/o8j5wiAsqSPoWELDUdJXhjAhsVliSn7bWZjOhrgQ= github.com/go-logr/logr v1.2.2/go.mod h1:jdQByPbusPIv2/zmleS9BjJVeZ6kBagPoEUsqbVz/1A= @@ -70,8 +64,6 @@ github.com/go-playground/validator/v10 v10.30.1 h1:f3zDSN/zOma+w6+1Wswgd9fLkdwy0 github.com/go-playground/validator/v10 v10.30.1/go.mod h1:oSuBIQzuJxL//3MelwSLD5hc2Tu889bF0Idm9Dg26cM= github.com/goccy/go-json v0.10.5 h1:Fq85nIqj+gXn/S5ahsiTlK3TmC85qgirsdTP/+DeaC4= github.com/goccy/go-json v0.10.5/go.mod h1:oq7eo15ShAhp70Anwd5lgX2pLfOS3QCiwU/PULtXL6M= -github.com/goccy/go-yaml v1.18.0 h1:8W7wMFS12Pcas7KU+VVkaiCng+kG8QiFeFwzFb+rwuw= -github.com/goccy/go-yaml v1.18.0/go.mod h1:XBurs7gK8ATbW4ZPGKgcbrY1Br56PdM69F7LkFRi1kA= github.com/goccy/go-yaml v1.19.2 h1:PmFC1S6h8ljIz6gMRBopkjP1TVT7xuwrButHID66PoM= github.com/goccy/go-yaml v1.19.2/go.mod h1:XBurs7gK8ATbW4ZPGKgcbrY1Br56PdM69F7LkFRi1kA= github.com/golang-jwt/jwt/v5 v5.3.1 h1:kYf81DTWFe7t+1VvL7eS+jKFVWaUnK9cB1qbwn63YCY= @@ -79,14 +71,16 @@ github.com/golang-jwt/jwt/v5 v5.3.1/go.mod h1:fxCRLWMO43lRc8nhHWY6LGqRcf+1gQWArs github.com/google/go-cmp v0.7.0 h1:wk8382ETsv4JYUZwIsn6YpYiWiBsYLSJiTsyBybVuN8= github.com/google/go-cmp v0.7.0/go.mod h1:pXiqmnSA92OHEEa9HXL2W4E7lf9JzCmGVUdgjX3N/iU= github.com/google/gofuzz v1.0.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg= -github.com/google/pprof v0.0.0-20221118152302-e6195bd50e26 h1:Xim43kblpZXfIBQsbuBVKCudVG457BR2GZFIz3uw3hQ= -github.com/google/pprof v0.0.0-20221118152302-e6195bd50e26/go.mod h1:dDKJzRmX4S37WGHujM7tX//fmj1uioxKzKxz3lo4HJo= +github.com/google/pprof v0.0.0-20250317173921-a4b03ec1a45e h1:ijClszYn+mADRFY17kjQEVQ1XRhq2/JR1M3sGqeJoxs= +github.com/google/pprof v0.0.0-20250317173921-a4b03ec1a45e/go.mod h1:boTsfXsheKC2y+lKOCMpSfarhxDeIzfZG1jqGcPl3cA= github.com/google/uuid v1.6.0 h1:NIvaJDMOsjHA8n1jAhLSgzrAzy1Hgr+hNrb57e+94F0= github.com/google/uuid v1.6.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo= github.com/gorilla/websocket v1.5.3 h1:saDtZ6Pbx/0u+bgYQ3q96pZgCzfhKXGPqt7kZ72aNNg= github.com/gorilla/websocket v1.5.3/go.mod h1:YR8l580nyteQvAITg2hZ9XVh4b55+EU/adAjf1fMHhE= github.com/grpc-ecosystem/grpc-gateway/v2 v2.27.2 h1:8Tjv8EJ+pM1xP8mK6egEbD1OgnVTyacbefKhmbLhIhU= github.com/grpc-ecosystem/grpc-gateway/v2 v2.27.2/go.mod h1:pkJQ2tZHJ0aFOVEEot6oZmaVEZcRme73eIFmhiVuRWs= +github.com/hashicorp/golang-lru/v2 v2.0.7 h1:a+bsQ5rvGLjzHuww6tVxozPZFVghXaHOwFs4luLUK2k= +github.com/hashicorp/golang-lru/v2 v2.0.7/go.mod h1:QeFd9opnmA6QUJc5vARoKUSoFhyfM2/ZepoAG6RGpeM= github.com/jinzhu/inflection v1.0.0 h1:K317FqzuhWc8YvSVlFMCCUb36O/S9MCKRDI7QkRKD/E= github.com/jinzhu/inflection v1.0.0/go.mod h1:h+uFLlag+Qp1Va5pdKtLDYj+kHp5pxUVkryuEj+Srlc= github.com/jinzhu/now v1.1.5 h1:/o9tlHleP7gOFmsnYNz3RGnqzefHA47wQpKrrdTIwXQ= @@ -126,6 +120,8 @@ github.com/morikuni/aec v1.0.0 h1:nP9CBfwrvYnBRgY6qfDQkygYDmYwOilePFkwzv4dU8A= github.com/morikuni/aec v1.0.0/go.mod h1:BbKIizmSmc5MMPqRYbxO4ZU0S0+P200+tUnFx7PXmsc= github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 h1:C3w9PqII01/Oq1c1nUAm88MOHcQC9l5mIlSMApZMrHA= github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822/go.mod h1:+n7T8mK8HuQTcFwEeznm/DIxMOiR9yIdICNftLE1DvQ= +github.com/ncruces/go-strftime v1.0.0 h1:HMFp8mLCTPp341M/ZnA4qaf7ZlsbTc+miZjCLOFAw7w= +github.com/ncruces/go-strftime v1.0.0/go.mod h1:Fwc5htZGVVkseilnfgOVb9mKy6w1naJmn9CehxcKcls= github.com/opencontainers/go-digest v1.0.0 h1:apOUWs51W5PlhuyGyz9FCeeBIOUDA/6nW8Oi/yOhh5U= github.com/opencontainers/go-digest v1.0.0/go.mod h1:0JzlMkj0TRzQZfJkVvzbP0HBR3IKzErnv2BNG4W4MAM= github.com/opencontainers/image-spec v1.1.1 h1:y0fUlFfIZhPF1W537XOLg0/fcx6zcHCJwooC2xJA040= @@ -144,21 +140,20 @@ github.com/prometheus/client_golang v1.23.2 h1:Je96obch5RDVy3FDMndoUsjAhG5Edi49h github.com/prometheus/client_golang v1.23.2/go.mod h1:Tb1a6LWHB3/SPIzCoaDXI4I8UHKeFTEQ1YCr+0Gyqmg= github.com/prometheus/client_model v0.6.2 h1:oBsgwpGs7iVziMvrGhE53c/GrLUsZdHnqNwqPLxwZyk= github.com/prometheus/client_model v0.6.2/go.mod h1:y3m2F6Gdpfy6Ut/GBsUqTWZqCUvMVzSfMLjcu6wAwpE= -github.com/prometheus/common v0.66.1 h1:h5E0h5/Y8niHc5DlaLlWLArTQI7tMrsfQjHV+d9ZoGs= -github.com/prometheus/common v0.66.1/go.mod h1:gcaUsgf3KfRSwHY4dIMXLPV0K/Wg1oZ8+SbZk/HH/dA= -github.com/prometheus/procfs v0.16.1 h1:hZ15bTNuirocR6u0JZ6BAHHmwS1p8B4P6MRqxtzMyRg= -github.com/prometheus/procfs v0.16.1/go.mod h1:teAbpZRB1iIAJYREa1LsoWUXykVXA1KlTmWl8x/U+Is= +github.com/prometheus/common v0.67.5 h1:pIgK94WWlQt1WLwAC5j2ynLaBRDiinoAb86HZHTUGI4= +github.com/prometheus/common v0.67.5/go.mod h1:SjE/0MzDEEAyrdr5Gqc6G+sXI67maCxzaT3A2+HqjUw= +github.com/prometheus/procfs v0.20.1 h1:XwbrGOIplXW/AU3YhIhLODXMJYyC1isLFfYCsTEycfc= +github.com/prometheus/procfs v0.20.1/go.mod h1:o9EMBZGRyvDrSPH1RqdxhojkuXstoe4UlK79eF5TGGo= github.com/quic-go/qpack v0.6.0 h1:g7W+BMYynC1LbYLSqRt8PBg5Tgwxn214ZZR34VIOjz8= github.com/quic-go/qpack v0.6.0/go.mod h1:lUpLKChi8njB4ty2bFLX2x4gzDqXwUpaO1DP9qMDZII= github.com/quic-go/quic-go v0.59.0 h1:OLJkp1Mlm/aS7dpKgTc6cnpynnD2Xg7C1pwL6vy/SAw= github.com/quic-go/quic-go v0.59.0/go.mod h1:upnsH4Ju1YkqpLXC305eW3yDZ4NfnNbmQRCMWS58IKU= -github.com/remyoudompheng/bigfft v0.0.0-20200410134404-eec4a21b6bb0/go.mod h1:qqbHyh8v60DhA7CoWK5oRCqLrMHRGoxYCSS9EjAz6Eo= github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec h1:W09IVJc94icq4NjY3clb7Lk8O1qJ8BdBEF8z0ibU0rE= github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec/go.mod h1:qqbHyh8v60DhA7CoWK5oRCqLrMHRGoxYCSS9EjAz6Eo= github.com/robfig/cron/v3 v3.0.1 h1:WdRxkvbJztn8LMz/QEvLN5sBU+xKpSqwwUO1Pjr4qDs= github.com/robfig/cron/v3 v3.0.1/go.mod h1:eQICP3HwyT7UooqI/z+Ov+PtYAWygg1TEWWzGIFLtro= -github.com/rogpeppe/go-internal v1.13.1 h1:KvO1DLK/DRN07sQ1LQKScxyZJuNnedQ5/wKSR38lUII= -github.com/rogpeppe/go-internal v1.13.1/go.mod h1:uMEvuHeurkdAXX61udpOXGD/AzZDWNMNyH2VO9fmH0o= +github.com/rogpeppe/go-internal v1.14.1 h1:UQB4HGPB6osV0SQTLymcB4TgvyWu6ZyliaW0tI/otEQ= +github.com/rogpeppe/go-internal v1.14.1/go.mod h1:MaRKkUm5W0goXpeCfT7UZI6fk/L7L7so1lCWt35ZSgc= github.com/sirupsen/logrus v1.9.4 h1:TsZE7l11zFCLZnZ+teH4Umoq5BhEIfIzfRDZ1Uzql2w= github.com/sirupsen/logrus v1.9.4/go.mod h1:ftWc9WdOfJ0a92nsE2jF5u5ZwH8Bv2zdeOC42RjbV2g= github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME= @@ -169,53 +164,52 @@ github.com/stretchr/objx v0.5.2/go.mod h1:FRsXN1f5AsAjCGJKqEizvkpNtU+EGNCLh3NxZ/ github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI= github.com/stretchr/testify v1.7.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg= github.com/stretchr/testify v1.8.0/go.mod h1:yNjHg4UonilssWZ8iaSj1OCr/vHnekPRkoO+kdMU+MU= -github.com/stretchr/testify v1.8.1/go.mod h1:w2LPCIKwWwSfY2zedu0+kehJoqGctiVI29o6fzry7u4= github.com/stretchr/testify v1.8.4/go.mod h1:sz/lmYIOXD/1dqDmKjjqLyZ2RngseejIcXlSw2iwfAo= github.com/stretchr/testify v1.10.0/go.mod h1:r2ic/lqez/lEtzL7wO/rwa5dbSLXVDPFyf8C91i36aY= github.com/stretchr/testify v1.11.1 h1:7s2iGBzp5EwR7/aIZr8ao5+dra3wiQyKjjFuvgVKu7U= github.com/stretchr/testify v1.11.1/go.mod h1:wZwfW3scLgRK+23gO65QZefKpKQRnfz6sD981Nm4B6U= github.com/twitchyliquid64/golang-asm v0.15.1 h1:SU5vSMR7hnwNxj24w34ZyCi/FmDZTkS4MhqMhdFk5YI= github.com/twitchyliquid64/golang-asm v0.15.1/go.mod h1:a1lVb/DtPvCB8fslRZhAngC2+aY1QWCk3Cedj/Gdt08= -github.com/ugorji/go/codec v1.3.0 h1:Qd2W2sQawAfG8XSvzwhBeoGq71zXOC/Q1E9y/wUcsUA= -github.com/ugorji/go/codec v1.3.0/go.mod h1:pRBVtBSKl77K30Bv8R2P+cLSGaTtex6fsA2Wjqmfxj4= github.com/ugorji/go/codec v1.3.1 h1:waO7eEiFDwidsBN6agj1vJQ4AG7lh2yqXyOXqhgQuyY= github.com/ugorji/go/codec v1.3.1/go.mod h1:pRBVtBSKl77K30Bv8R2P+cLSGaTtex6fsA2Wjqmfxj4= go.mongodb.org/mongo-driver/v2 v2.5.0 h1:yXUhImUjjAInNcpTcAlPHiT7bIXhshCTL3jVBkF3xaE= go.mongodb.org/mongo-driver/v2 v2.5.0/go.mod h1:yOI9kBsufol30iFsl1slpdq1I0eHPzybRWdyYUs8K/0= -go.opentelemetry.io/auto/sdk v1.1.0 h1:cH53jehLUN6UFLY71z+NDOiNJqDdPRaXzTel0sJySYA= -go.opentelemetry.io/auto/sdk v1.1.0/go.mod h1:3wSPjt5PWp2RhlCcmmOial7AvC4DQqZb7a7wCow3W8A= -go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.63.0 h1:RbKq8BG0FI8OiXhBfcRtqqHcZcka+gU3cskNuf05R18= -go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.63.0/go.mod h1:h06DGIukJOevXaj/xrNjhi/2098RZzcLTbc0jDAUbsg= -go.opentelemetry.io/otel v1.38.0 h1:RkfdswUDRimDg0m2Az18RKOsnI8UDzppJAtj01/Ymk8= -go.opentelemetry.io/otel v1.38.0/go.mod h1:zcmtmQ1+YmQM9wrNsTGV/q/uyusom3P8RxwExxkZhjM= +go.opentelemetry.io/auto/sdk v1.2.1 h1:jXsnJ4Lmnqd11kwkBV2LgLoFMZKizbCi5fNZ/ipaZ64= +go.opentelemetry.io/auto/sdk v1.2.1/go.mod h1:KRTj+aOaElaLi+wW1kO/DZRXwkF4C5xPbEe3ZiIhN7Y= +go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.65.0 h1:7iP2uCb7sGddAr30RRS6xjKy7AZ2JtTOPA3oolgVSw8= +go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.65.0/go.mod h1:c7hN3ddxs/z6q9xwvfLPk+UHlWRQyaeR1LdgfL/66l0= +go.opentelemetry.io/otel v1.40.0 h1:oA5YeOcpRTXq6NN7frwmwFR0Cn3RhTVZvXsP4duvCms= +go.opentelemetry.io/otel v1.40.0/go.mod h1:IMb+uXZUKkMXdPddhwAHm6UfOwJyh4ct1ybIlV14J0g= go.opentelemetry.io/otel/exporters/otlp/otlptrace v1.38.0 h1:GqRJVj7UmLjCVyVJ3ZFLdPRmhDUp2zFmQe3RHIOsw24= go.opentelemetry.io/otel/exporters/otlp/otlptrace v1.38.0/go.mod h1:ri3aaHSmCTVYu2AWv44YMauwAQc0aqI9gHKIcSbI1pU= go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.38.0 h1:aTL7F04bJHUlztTsNGJ2l+6he8c+y/b//eR0jjjemT4= go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.38.0/go.mod h1:kldtb7jDTeol0l3ewcmd8SDvx3EmIE7lyvqbasU3QC4= -go.opentelemetry.io/otel/metric v1.38.0 h1:Kl6lzIYGAh5M159u9NgiRkmoMKjvbsKtYRwgfrA6WpA= -go.opentelemetry.io/otel/metric v1.38.0/go.mod h1:kB5n/QoRM8YwmUahxvI3bO34eVtQf2i4utNVLr9gEmI= -go.opentelemetry.io/otel/sdk v1.38.0 h1:l48sr5YbNf2hpCUj/FoGhW9yDkl+Ma+LrVl8qaM5b+E= -go.opentelemetry.io/otel/sdk v1.38.0/go.mod h1:ghmNdGlVemJI3+ZB5iDEuk4bWA3GkTpW+DOoZMYBVVg= -go.opentelemetry.io/otel/sdk/metric v1.38.0 h1:aSH66iL0aZqo//xXzQLYozmWrXxyFkBJ6qT5wthqPoM= -go.opentelemetry.io/otel/sdk/metric v1.38.0/go.mod h1:dg9PBnW9XdQ1Hd6ZnRz689CbtrUp0wMMs9iPcgT9EZA= -go.opentelemetry.io/otel/trace v1.38.0 h1:Fxk5bKrDZJUH+AMyyIXGcFAPah0oRcT+LuNtJrmcNLE= -go.opentelemetry.io/otel/trace v1.38.0/go.mod h1:j1P9ivuFsTceSWe1oY+EeW3sc+Pp42sO++GHkg4wwhs= +go.opentelemetry.io/otel/metric v1.40.0 h1:rcZe317KPftE2rstWIBitCdVp89A2HqjkxR3c11+p9g= +go.opentelemetry.io/otel/metric v1.40.0/go.mod h1:ib/crwQH7N3r5kfiBZQbwrTge743UDc7DTFVZrrXnqc= +go.opentelemetry.io/otel/sdk v1.40.0 h1:KHW/jUzgo6wsPh9At46+h4upjtccTmuZCFAc9OJ71f8= +go.opentelemetry.io/otel/sdk v1.40.0/go.mod h1:Ph7EFdYvxq72Y8Li9q8KebuYUr2KoeyHx0DRMKrYBUE= +go.opentelemetry.io/otel/sdk/metric v1.40.0 h1:mtmdVqgQkeRxHgRv4qhyJduP3fYJRMX4AtAlbuWdCYw= +go.opentelemetry.io/otel/sdk/metric v1.40.0/go.mod h1:4Z2bGMf0KSK3uRjlczMOeMhKU2rhUqdWNoKcYrtcBPg= +go.opentelemetry.io/otel/trace v1.40.0 h1:WA4etStDttCSYuhwvEa8OP8I5EWu24lkOzp+ZYblVjw= +go.opentelemetry.io/otel/trace v1.40.0/go.mod h1:zeAhriXecNGP/s2SEG3+Y8X9ujcJOTqQ5RgdEJcawiA= go.opentelemetry.io/proto/otlp v1.7.1 h1:gTOMpGDb0WTBOP8JaO72iL3auEZhVmAQg4ipjOVAtj4= go.opentelemetry.io/proto/otlp v1.7.1/go.mod h1:b2rVh6rfI/s2pHWNlB7ILJcRALpcNDzKhACevjI+ZnE= go.uber.org/goleak v1.3.0 h1:2K3zAYmnTNqV73imy9J1T3WC+gmCePx2hEGkimedGto= go.uber.org/goleak v1.3.0/go.mod h1:CoHD4mav9JJNrW/WLlf7HGZPjdw8EucARQHekz1X6bE= go.uber.org/mock v0.6.0 h1:hyF9dfmbgIX5EfOdasqLsWD6xqpNZlXblLB/Dbnwv3Y= go.uber.org/mock v0.6.0/go.mod h1:KiVJ4BqZJaMj4svdfmHM0AUx4NJYO8ZNpPnZn1Z+BBU= -go.yaml.in/yaml/v2 v2.4.2 h1:DzmwEr2rDGHl7lsFgAHxmNz/1NlQ7xLIrlN2h5d1eGI= -go.yaml.in/yaml/v2 v2.4.2/go.mod h1:081UH+NErpNdqlCXm3TtEran0rJZGxAYx9hb/ELlsPU= -golang.org/x/arch v0.22.0 h1:c/Zle32i5ttqRXjdLyyHZESLD/bB90DCU1g9l/0YBDI= -golang.org/x/arch v0.22.0/go.mod h1:dNHoOeKiyja7GTvF9NJS1l3Z2yntpQNzgrjh1cU103A= +go.yaml.in/yaml/v2 v2.4.3 h1:6gvOSjQoTB3vt1l+CU+tSyi/HOjfOjRLJ4YwYZGwRO0= +go.yaml.in/yaml/v2 v2.4.3/go.mod h1:zSxWcmIDjOzPXpjlTTbAsKokqkDNAVtZO0WOMiT90s8= +golang.org/x/arch v0.24.0 h1:qlJ3M9upxvFfwRM51tTg3Yl+8CP9vCC1E7vlFpgv99Y= +golang.org/x/arch v0.24.0/go.mod h1:dNHoOeKiyja7GTvF9NJS1l3Z2yntpQNzgrjh1cU103A= golang.org/x/crypto v0.48.0 h1:/VRzVqiRSggnhY7gNRxPauEQ5Drw9haKdM0jqfcCFts= golang.org/x/crypto v0.48.0/go.mod h1:r0kV5h3qnFPlQnBSrULhlsRfryS2pmewsg+XfMgkVos= -golang.org/x/net v0.50.0 h1:ucWh9eiCGyDR3vtzso0WMQinm2Dnt8cFMuQa9K33J60= -golang.org/x/net v0.50.0/go.mod h1:UgoSli3F/pBgdJBHCTc+tp3gmrU4XswgGRgtnwWTfyM= +golang.org/x/mod v0.32.0 h1:9F4d3PHLljb6x//jOyokMv3eX+YDeepZSEo3mFJy93c= +golang.org/x/mod v0.32.0/go.mod h1:SgipZ/3h2Ci89DlEtEXWUk/HteuRin+HHhN+WbNhguU= golang.org/x/net v0.51.0 h1:94R/GTO7mt3/4wIKpcR5gkGmRLOuE/2hNGeWq/GBIFo= golang.org/x/net v0.51.0/go.mod h1:aamm+2QF5ogm02fjy5Bb7CQ0WMt1/WVM7FtyaTLlA9Y= +golang.org/x/sync v0.19.0 h1:vV+1eWNmZ5geRlYjzm2adRgW2/mcpevXNg50YZtPCE4= +golang.org/x/sync v0.19.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI= golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg= golang.org/x/sys v0.41.0 h1:Ivj+2Cp/ylzLiEU89QhWblYnOE9zerudt9Ftecq2C6k= golang.org/x/sys v0.41.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks= @@ -223,6 +217,8 @@ golang.org/x/text v0.34.0 h1:oL/Qq0Kdaqxa1KbNeMKwQq0reLCCaFtqu2eNuSeNHbk= golang.org/x/text v0.34.0/go.mod h1:homfLqTYRFyVYemLBFl5GgL/DWEiH5wcsQ5gSh1yziA= golang.org/x/time v0.14.0 h1:MRx4UaLrDotUKUdCIqzPC48t1Y9hANFKIRpNx+Te8PI= golang.org/x/time v0.14.0/go.mod h1:eL/Oa2bBBK0TkX57Fyni+NgnyQQN4LitPmob2Hjnqw4= +golang.org/x/tools v0.41.0 h1:a9b8iMweWG+S0OBnlU36rzLp20z1Rp10w+IY2czHTQc= +golang.org/x/tools v0.41.0/go.mod h1:XSY6eDqxVNiYgezAVqqCeihT4j1U2CCsqvH3WhQpnlg= google.golang.org/genproto/googleapis/api v0.0.0-20250825161204-c5933d9347a5 h1:BIRfGDEjiHRrk0QKZe3Xv2ieMhtgRGeLcZQ0mIVn4EY= google.golang.org/genproto/googleapis/api v0.0.0-20250825161204-c5933d9347a5/go.mod h1:j3QtIyytwqGr1JUDtYXwtMXWPKsEa5LtzIFN1Wn5WvE= google.golang.org/genproto/googleapis/rpc v0.0.0-20250825161204-c5933d9347a5 h1:eaY8u2EuxbRv7c3NiGK0/NedzVsCcV6hDuU5qPX5EGE= @@ -245,11 +241,31 @@ gorm.io/gorm v1.31.1 h1:7CA8FTFz/gRfgqgpeKIBcervUn3xSyPUmr6B2WXJ7kg= gorm.io/gorm v1.31.1/go.mod h1:XyQVbO2k6YkOis7C2437jSit3SsDK72s7n7rsSHd+Gs= gotest.tools/v3 v3.5.2 h1:7koQfIKdy+I8UTetycgUqXWSDwpgv193Ka+qRsmBY8Q= gotest.tools/v3 v3.5.2/go.mod h1:LtdLGcnqToBH83WByAAi/wiwSFCArdFIUV/xxN4pcjA= -modernc.org/libc v1.22.5 h1:91BNch/e5B0uPbJFgqbxXuOnxBQjlS//icfQEGmvyjE= -modernc.org/libc v1.22.5/go.mod h1:jj+Z7dTNX8fBScMVNRAYZ/jF91K8fdT2hYMThc3YjBY= -modernc.org/mathutil v1.5.0 h1:rV0Ko/6SfM+8G+yKiyI830l3Wuz1zRutdslNoQ0kfiQ= -modernc.org/mathutil v1.5.0/go.mod h1:mZW8CKdRPY1v87qxC/wUdX5O1qDzXMP5TH3wjfpga6E= -modernc.org/memory v1.5.0 h1:N+/8c5rE6EqugZwHii4IFsaJ7MUhoWX07J5tC/iI5Ds= -modernc.org/memory v1.5.0/go.mod h1:PkUhL0Mugw21sHPeskwZW4D6VscE/GQJOnIpCnW6pSU= -modernc.org/sqlite v1.23.1 h1:nrSBg4aRQQwq59JpvGEQ15tNxoO5pX/kUjcRNwSAGQM= -modernc.org/sqlite v1.23.1/go.mod h1:OrDj17Mggn6MhE+iPbBNf7RGKODDE9NFT0f3EwDzJqk= +modernc.org/cc/v4 v4.27.1 h1:9W30zRlYrefrDV2JE2O8VDtJ1yPGownxciz5rrbQZis= +modernc.org/cc/v4 v4.27.1/go.mod h1:uVtb5OGqUKpoLWhqwNQo/8LwvoiEBLvZXIQ/SmO6mL0= +modernc.org/ccgo/v4 v4.31.0 h1:/bsaxqdgX3gy/0DboxcvWrc3NpzH+6wpFfI/ZaA/hrg= +modernc.org/ccgo/v4 v4.31.0/go.mod h1:jKe8kPBjIN/VdGTVqARTQ8N1gAziBmiISY8j5HoKwjg= +modernc.org/fileutil v1.4.0 h1:j6ZzNTftVS054gi281TyLjHPp6CPHr2KCxEXjEbD6SM= +modernc.org/fileutil v1.4.0/go.mod h1:EqdKFDxiByqxLk8ozOxObDSfcVOv/54xDs/DUHdvCUU= +modernc.org/gc/v2 v2.6.5 h1:nyqdV8q46KvTpZlsw66kWqwXRHdjIlJOhG6kxiV/9xI= +modernc.org/gc/v2 v2.6.5/go.mod h1:YgIahr1ypgfe7chRuJi2gD7DBQiKSLMPgBQe9oIiito= +modernc.org/gc/v3 v3.1.2 h1:ZtDCnhonXSZexk/AYsegNRV1lJGgaNZJuKjJSWKyEqo= +modernc.org/gc/v3 v3.1.2/go.mod h1:HFK/6AGESC7Ex+EZJhJ2Gni6cTaYpSMmU/cT9RmlfYY= +modernc.org/goabi0 v0.2.0 h1:HvEowk7LxcPd0eq6mVOAEMai46V+i7Jrj13t4AzuNks= +modernc.org/goabi0 v0.2.0/go.mod h1:CEFRnnJhKvWT1c1JTI3Avm+tgOWbkOu5oPA8eH8LnMI= +modernc.org/libc v1.68.1 h1:qNL/EzzdzNicXwJ9Gj2IHlVjuqRQsPXngFRaDMGuFwE= +modernc.org/libc v1.68.1/go.mod h1:YfLLduUEbodNV2xLU5JOnRHBTAHVHsVW3bVYGw0ZCV4= +modernc.org/mathutil v1.7.1 h1:GCZVGXdaN8gTqB1Mf/usp1Y/hSqgI2vAGGP4jZMCxOU= +modernc.org/mathutil v1.7.1/go.mod h1:4p5IwJITfppl0G4sUEDtCr4DthTaT47/N3aT6MhfgJg= +modernc.org/memory v1.11.0 h1:o4QC8aMQzmcwCK3t3Ux/ZHmwFPzE6hf2Y5LbkRs+hbI= +modernc.org/memory v1.11.0/go.mod h1:/JP4VbVC+K5sU2wZi9bHoq2MAkCnrt2r98UGeSK7Mjw= +modernc.org/opt v0.1.4 h1:2kNGMRiUjrp4LcaPuLY2PzUfqM/w9N23quVwhKt5Qm8= +modernc.org/opt v0.1.4/go.mod h1:03fq9lsNfvkYSfxrfUhZCWPk1lm4cq4N+Bh//bEtgns= +modernc.org/sortutil v1.2.1 h1:+xyoGf15mM3NMlPDnFqrteY07klSFxLElE2PVuWIJ7w= +modernc.org/sortutil v1.2.1/go.mod h1:7ZI3a3REbai7gzCLcotuw9AC4VZVpYMjDzETGsSMqJE= +modernc.org/sqlite v1.46.1 h1:eFJ2ShBLIEnUWlLy12raN0Z1plqmFX9Qe3rjQTKt6sU= +modernc.org/sqlite v1.46.1/go.mod h1:CzbrU2lSB1DKUusvwGz7rqEKIq+NUd8GWuBBZDs9/nA= +modernc.org/strutil v1.2.1 h1:UneZBkQA+DX2Rp35KcM69cSsNES9ly8mQWD71HKlOA0= +modernc.org/strutil v1.2.1/go.mod h1:EHkiggD70koQxjVdSBM3JKM7k6L0FbGE5eymy9i3B9A= +modernc.org/token v1.1.0 h1:Xl7Ap9dKaEs5kLoOQeQmPWevfnk/DM5qcLcYlA8ys6Y= +modernc.org/token v1.1.0/go.mod h1:UGzOrNV1mAFSEB63lOFHIpNRUVMvYTc6yu1SMY/XTDM= From d9cc0ead7180bfa8a178eb0e6f4b98e1d6a5f911 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 01:43:10 +0000 Subject: [PATCH 06/38] chore: move ACL and Security Headers hotfix plan documentation to archive --- .../acl_security_headers_hotfix_plan.md | 270 ++++++++++++++++++ 1 file changed, 270 insertions(+) create mode 100644 docs/plans/archive/acl_security_headers_hotfix_plan.md diff --git a/docs/plans/archive/acl_security_headers_hotfix_plan.md b/docs/plans/archive/acl_security_headers_hotfix_plan.md new file mode 100644 index 00000000..81fc1c46 --- /dev/null +++ b/docs/plans/archive/acl_security_headers_hotfix_plan.md @@ -0,0 +1,270 @@ +# ACL + Security Headers Hotfix Plan (Proxy Host Create/Edit) + +## 1. Introduction + +### Overview +Hotfix request: Proxy Host form dropdown selections for Access Control List (ACL) and Security Headers are not being applied/persisted for new or edited hosts. + +Reported behavior: +1. Existing hosts with previously assigned ACL/Security Header profile retain old values. +2. Users cannot reliably remove or change those values in UI. +3. Newly created hosts cannot reliably apply ACL/Security Header profile. + +### Objective +Deliver an urgent but correct root-cause fix across frontend binding and backend persistence flow, with minimum user interruption and full validation gates. + +## 2. Research Findings (Current Architecture + Touchpoints) + +### Frontend Entry Points +1. `frontend/src/pages/ProxyHosts.tsx` + - `handleSubmit(data)` calls `updateHost(editingHost.uuid, data)` or `createHost(data)`. + - Renders `ProxyHostForm` modal for create/edit flows. +2. `frontend/src/components/ProxyHostForm.tsx` + - Local form state initializes `access_list_id` and `security_header_profile_id`. + - ACL control uses `AccessListSelector`. + - Security Headers control uses `Select` with `security_header_profile_id` mapping. + - Submission path: `handleSubmit` -> `onSubmit(payloadWithoutUptime)`. +3. `frontend/src/components/AccessListSelector.tsx` + - Converts select values between `string` and `number | null`. + +### Frontend API/Hooks +1. `frontend/src/hooks/useProxyHosts.ts` + - `createHost` -> `createProxyHost`. + - `updateHost` -> `updateProxyHost`. +2. `frontend/src/api/proxyHosts.ts` + - `createProxyHost(host: Partial)` -> `POST /api/v1/proxy-hosts`. + - `updateProxyHost(uuid, host)` -> `PUT /api/v1/proxy-hosts/:uuid`. + - Contract fields: `access_list_id`, `security_header_profile_id`. + +### Backend Entry/Transformation/Persistence +1. Route registration + - `backend/internal/api/routes/routes.go`: `proxyHostHandler.RegisterRoutes(protected)`. +2. Handler + - `backend/internal/api/handlers/proxy_host_handler.go` + - `Create(c)` uses `ShouldBindJSON(&models.ProxyHost{})`. + - `Update(c)` uses `map[string]any` partial update parsing. + - Target fields: + - `payload["access_list_id"]` -> `parseNullableUintField` -> `host.AccessListID` + - `payload["security_header_profile_id"]` -> typed conversion -> `host.SecurityHeaderProfileID` +3. Service + - `backend/internal/services/proxyhost_service.go` + - `Create(host)` validates + `db.Create(host)`. + - `Update(host)` validates + `db.Model(...).Select("*").Updates(host)`. +4. Model + - `backend/internal/models/proxy_host.go` + - `AccessListID *uint \`json:"access_list_id"\`` + - `SecurityHeaderProfileID *uint \`json:"security_header_profile_id"\`` + +### Existing Tests Relevant to Incident +1. Frontend unit regression coverage already exists: + - `frontend/src/components/__tests__/ProxyHostForm-dropdown-changes.test.tsx` +2. E2E regression spec exists: + - `tests/security-enforcement/acl-dropdown-regression.spec.ts` +3. Backend update and security-header tests exist: + - `backend/internal/api/handlers/proxy_host_handler_update_test.go` + - `backend/internal/api/handlers/proxy_host_handler_security_headers_test.go` + +## 3. Root-Cause-First Trace + +### Trace Model (Mandatory) +1. Entry Point: + - UI dropdown interactions in `ProxyHostForm` and `AccessListSelector`. +2. Transformation: + - Form state conversion (`string` <-> `number | null`) and payload construction in `ProxyHostForm`. + - API serialization via `frontend/src/api/proxyHosts.ts`. +3. Persistence: + - Backend `Update` parser (`proxy_host_handler.go`) and `ProxyHostService.Update` persistence. +4. Exit Point: + - Response body consumed by React Query invalidation/refetch in `useProxyHosts`. + - UI reflects updated values in table/form. + +### Most Likely Failure Zones +1. Frontend select binding/conversion drift (top candidate) + - Shared symptom across ACL and Security Headers points to form/select layer. + - Candidate files: + - `frontend/src/components/ProxyHostForm.tsx` + - `frontend/src/components/AccessListSelector.tsx` + - `frontend/src/components/ui/Select.tsx` +2. Payload mutation or stale form object behavior + - Ensure payload carries updated `access_list_id` / `security_header_profile_id` values at submit time. +3. Backend partial-update parser edge behavior + - Ensure `nil`, numeric string, and number conversions are consistent between ACL and security header profile paths. + +### Investigation Decision +Root-cause verification will be instrumented through failing-first Playwright scenario and targeted handler tests before applying code changes. + +## 4. EARS Requirements + +1. WHEN a user selects an ACL in the Proxy Host create/edit form, THE SYSTEM SHALL persist `access_list_id` and return it in API response. +2. WHEN a user changes ACL from one value to another, THE SYSTEM SHALL replace prior `access_list_id` with the new value. +3. WHEN a user selects "No Access Control", THE SYSTEM SHALL persist `access_list_id = null`. +4. WHEN a user selects a Security Headers profile in the Proxy Host create/edit form, THE SYSTEM SHALL persist `security_header_profile_id` and return it in API response. +5. WHEN a user changes Security Headers profile from one value to another, THE SYSTEM SHALL replace prior `security_header_profile_id` with the new value. +6. WHEN a user selects "None" for Security Headers, THE SYSTEM SHALL persist `security_header_profile_id = null`. +7. IF dropdown interaction fails to update internal form state, THEN THE SYSTEM SHALL prevent stale values from being persisted. +8. WHILE updating Proxy Host settings, THE SYSTEM SHALL maintain existing behavior for unrelated fields and not regress certificate, DNS challenge, or uptime-linked updates. + +Note: User-visible blocking error behavior is deferred unless required by confirmed root cause. + +## 5. Technical Specification (Hotfix Scope) + +### API Contract (No Breaking Change) +1. `POST /api/v1/proxy-hosts` + - Request fields include `access_list_id`, `security_header_profile_id` as nullable numeric fields. +2. `PUT /api/v1/proxy-hosts/:uuid` + - Partial payload accepts nullable updates for both fields. +3. Response must echo persisted values in snake_case: + - `access_list_id` + - `security_header_profile_id` + +### Data Model/DB +No schema migration expected. Existing nullable FK fields in `backend/internal/models/proxy_host.go` are sufficient. + +### Targeted Code Areas for Fix +1. Frontend + - `frontend/src/components/ProxyHostForm.tsx` + - `frontend/src/components/AccessListSelector.tsx` + - `frontend/src/components/ui/Select.tsx` (only if click/select propagation issue confirmed) + - `frontend/src/api/proxyHosts.ts` (only if serialization issue confirmed) +2. Backend + - `backend/internal/api/handlers/proxy_host_handler.go` (only if parsing/persistence mismatch confirmed) + - `backend/internal/services/proxyhost_service.go` (only if update write path proves incorrect) + +## 6. Edge Cases + +1. Edit host with existing ACL/profile and switch to another value. +2. Edit host with existing ACL/profile and clear to null. +3. Create new host with ACL/profile set before first save. +4. Submit with stringified numeric values (defensive compatibility). +5. Submit with null values for both fields simultaneously. +6. Missing/deleted profile or ACL IDs in backend (validation errors). +7. Multiple rapid dropdown changes before save (last selection wins). + +## 7. Risk Analysis + +### High Risk +1. Silent stale-state submission from form controls. +2. Regressing other Proxy Host settings due to broad payload mutation. + +### Medium Risk +1. Partial-update parser divergence between ACL and security profile behavior. +2. UI select portal/z-index interaction causing non-deterministic click handling. + +### Mitigations +1. Reproduce with Playwright first and capture exact failing action path. +2. Add/strengthen focused frontend tests around create/edit/clear flows. +3. Add/strengthen backend tests for nullable + conversion paths. +4. Keep hotfix minimal and avoid unrelated refactors. + +## 8. Implementation Plan (Urgent, Minimal Interruption) + +### Phase 1: Reproduction + Guardrails (Playwright First) +1. Execute targeted E2E spec for dropdown flow and create/edit persistence behavior. +2. Capture exact failure step and confirm whether failure is click binding, payload value, or backend persistence. +3. Add/adjust failing-first test if current suite does not capture observed production regression. + +### Phase 2: Frontend Fix +1. Patch select binding/state mapping for ACL and Security Headers in `ProxyHostForm`/`AccessListSelector`. +2. If needed, patch `ui/Select` interaction layering. +3. Ensure payload contains correct final `access_list_id` and `security_header_profile_id` values at submit. +4. Extend `ProxyHostForm` tests for create/edit/change/remove flows. + +### Phase 3: Backend Hardening (Conditional) +1. Only if frontend payload is correct but persistence is wrong: + - Backend fix MUST use field-scoped partial-update semantics for `access_list_id` and `security_header_profile_id` only (unless separately justified). + - Ensure write path persists null transitions reliably. +2. Add/adjust handler/service regression tests proving no unintended mutation of unrelated proxy host fields during these targeted updates. + +### Phase 4: Integration + Regression +1. Run complete targeted Proxy Host UI flow tests. +2. Validate list refresh and modal reopen reflect persisted values. +3. Validate no regressions in bulk ACL / bulk security-header operations. + +### Phase 5: Documentation + Handoff +1. Update changelog/release notes only for hotfix behavior. +2. Keep architecture docs unchanged unless root cause requires architectural note. +3. Handoff to Supervisor agent for review after plan approval and implementation. + +## 9. Acceptance Criteria + +1. ACL dropdown selection persists on create and edit. +2. Security Headers dropdown selection persists on create and edit. +3. Clearing ACL persists `null` and is reflected after reload. +4. Clearing Security Headers persists `null` and is reflected after reload. +5. Existing hosts can change from one ACL/profile to another without stale value retention. +6. New hosts can apply ACL/profile at creation time. +7. No regressions in unrelated proxy host fields. +8. All validation gates in Section 11 pass. +9. API create response returns persisted `access_list_id` and `security_header_profile_id` matching submitted values (including `null`). +10. API update response returns persisted `access_list_id` and `security_header_profile_id` after `value->value`, `value->null`, and `null->value` transitions. +11. Backend persistence verification confirms unrelated proxy host fields remain unchanged for targeted updates. + +## 10. PR Slicing Strategy + +### Decision +Single PR (hotfix-first), with contingency split only if backend root cause is confirmed late. + +### Rationale +1. Incident impact is immediate user-facing and concentrated in one feature path. +2. Frontend + targeted backend/test changes are tightly coupled for verification. +3. Single PR minimizes release coordination and user interruption. + +### Contingency (Only if split becomes necessary) +1. PR-1: Frontend binding + tests + - Scope: `ProxyHostForm`, `AccessListSelector`, `ui/Select` (if required), related tests. + - Dependency: none. + - Acceptance: UI submit payload verified correct in unit + Playwright. +2. PR-2: Backend parser/persistence + tests (conditional) + - Scope: `proxy_host_handler.go`, `proxyhost_service.go`, handler/service tests. + - Dependency: PR-1 merged or rebased for aligned contract. + - Acceptance: API update/create persist both nullable IDs correctly. +3. PR-3: Regression hardening + docs + - Scope: extra regression coverage, release-note hotfix entry. + - Dependency: PR-1/PR-2. + - Acceptance: full DoD validation sequence passes. + +## 11. Validation Plan (Mandatory Sequence) + +0. E2E environment prerequisite + - Determine rebuild necessity per testing policy: if application/runtime or Docker input changes are present, rebuild is required. + - If rebuild is required or the container is unhealthy, run `.github/skills/scripts/skill-runner.sh docker-rebuild-e2e`. + - Record container health outcome before executing tests. +1. Playwright first + - Run targeted Proxy Host dropdown and create/edit persistence scenarios. +2. Local patch coverage preflight + - Generate `test-results/local-patch-report.md` and `test-results/local-patch-report.json`. +3. Unit and coverage + - Backend coverage run (threshold >= 85%). + - Frontend coverage run (threshold >= 85%). +4. Type checks + - Frontend TypeScript check. +5. Pre-commit + - `pre-commit run --all-files` with zero blocking failures. +6. Security scans + - CodeQL Go + JS (security-and-quality). + - Findings check gate. + - Trivy scan. + - Conditional GORM security scan if model/DB-layer changes are made. +7. Build verification + - Backend build + frontend build pass. + +## 12. File Review: `.gitignore`, `codecov.yml`, `.dockerignore`, `Dockerfile` + +Assessment for this hotfix: +1. `.gitignore`: no required change for ACL/Security Headers hotfix. +2. `codecov.yml`: no required change; current exclusions/thresholds are compatible. +3. `.dockerignore`: no required change unless new hotfix-only artifact paths are introduced. +4. `Dockerfile`: no required change; incident is application logic/UI binding, not image build pipeline. + +If implementation introduces new persistent test artifacts, update ignore files in the same PR. + +## 13. Rollback and Contingency + +1. If hotfix causes regression in proxy host save flow, revert hotfix commit and redeploy prior stable build. +2. If frontend-only fix is insufficient, activate conditional backend phase immediately. +3. If validation gates fail on security/coverage, hold merge until fixed; no partial exception for this incident. +4. Post-rollback smoke checks: + - Create host with ACL/profile. + - Edit to different ACL/profile values. + - Clear both values to `null`. + - Verify persisted values in API response and after UI reload. From 61d4e12c56078e80a573cbbccfb1b592531ded7e Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 02:06:46 +0000 Subject: [PATCH 07/38] fix(deps): update go.mod entries for various dependencies --- go.work.sum | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/go.work.sum b/go.work.sum index ebcc3c1e..468746d5 100644 --- a/go.work.sum +++ b/go.work.sum @@ -6,6 +6,7 @@ github.com/alecthomas/kingpin/v2 v2.4.0 h1:f48lwail6p8zpO1bC4TxtqACaGqHYA22qkHjH github.com/alecthomas/kingpin/v2 v2.4.0/go.mod h1:0gyi0zQnjuFk8xrkNKamJoyUo382HRL7ATRpFZCw6tE= github.com/alecthomas/units v0.0.0-20211218093645-b94a6e3cc137 h1:s6gZFSlWYmbqAuRjVTiNNhvNRfY2Wxp9nhfyel4rklc= github.com/alecthomas/units v0.0.0-20211218093645-b94a6e3cc137/go.mod h1:OMCwj8VM1Kc9e19TLln2VL61YJF0x1XFtfdL4JdbSyE= +github.com/alecthomas/units v0.0.0-20240927000941-0f3dac36c52b/go.mod h1:fvzegU4vN3H1qMT+8wDmzjAcDONcgo2/SZ/TyfdUOFs= github.com/cloudwego/iasm v0.2.0 h1:1KNIy1I1H9hNNFEEH3DVnI4UujN+1zjpuk6gwHLTssg= github.com/cloudwego/iasm v0.2.0/go.mod h1:8rXZaNYT2n95jn+zTI1sDr+IgcD2GVs0nlbbQPiEFhY= github.com/containerd/typeurl/v2 v2.2.0 h1:6NBDbQzr7I5LHgp34xAXYF5DOTQDn05X58lsPEmzLso= @@ -79,23 +80,22 @@ github.com/youmark/pkcs8 v0.0.0-20240726163527-a2c0da244d78/go.mod h1:aL8wCCfTfS github.com/yuin/goldmark v1.4.13 h1:fVcFKWvrslecOb/tg+Cc05dkeYx540o0FuFt3nUVDoE= github.com/yuin/goldmark v1.4.13/go.mod h1:6yULJ656Px+3vBD8DxQVa3kxgyrAnzto9xy5taEt/CY= golang.org/x/crypto v0.44.0/go.mod h1:013i+Nw79BMiQiMsOPcVCB5ZIJbYkerPrGnOa00tvmc= +golang.org/x/exp v0.0.0-20251023183803-a4bb9ffd2546/go.mod h1:j/pmGrbnkbPtQfxEe5D0VQhZC6qKbfKifgD0oM7sR70= golang.org/x/mod v0.18.0/go.mod h1:hTbmBsO62+eylJbnUtE2MGJUyE7QWk4xUqPFrRgJ+7c= golang.org/x/mod v0.29.0/go.mod h1:NyhrlYXJ2H4eJiRy/WDBO6HMqZQ6q9nk4JzS3NuCK+w= golang.org/x/mod v0.30.0 h1:fDEXFVZ/fmCKProc/yAXXUijritrDzahmwwefnjoPFk= golang.org/x/mod v0.30.0/go.mod h1:lAsf5O2EvJeSFMiBxXDki7sCgAxEUcZHXoXMKT4GJKc= golang.org/x/mod v0.31.0/go.mod h1:43JraMp9cGx1Rx3AqioxrbrhNsLl2l/iNAvuBkrezpg= -golang.org/x/mod v0.32.0/go.mod h1:SgipZ/3h2Ci89DlEtEXWUk/HteuRin+HHhN+WbNhguU= golang.org/x/net v0.43.0/go.mod h1:vhO1fvI4dGsIjh73sWfUVjj3N7CA9WkKJNQm2svM6Jg= golang.org/x/net v0.46.0/go.mod h1:Q9BGdFy1y4nkUwiLvT5qtyhAnEHgnQ/zd8PfU6nc210= golang.org/x/oauth2 v0.6.0 h1:Lh8GPgSKBfWSwFvtuWOfeI3aAAnbXTSutYxJiOJFgIw= golang.org/x/oauth2 v0.6.0/go.mod h1:ycmewcwgD4Rpr3eZJLSB4Kyyljb3qDh40vJ8STE5HKw= golang.org/x/oauth2 v0.30.0 h1:dnDm7JmhM45NNpd8FDDeLhK6FwqbOf4MLCM9zb1BOHI= golang.org/x/oauth2 v0.30.0/go.mod h1:B++QgG3ZKulg6sRPGD/mqlHQs5rB3Ml9erfeDY7xKlU= +golang.org/x/oauth2 v0.34.0/go.mod h1:lzm5WQJQwKZ3nwavOZ3IS5Aulzxi68dUSgRHujetwEA= golang.org/x/sync v0.13.0/go.mod h1:1dzgHSNfp02xaA81J2MS99Qcpr2w7fw1gpm99rleRqA= golang.org/x/sync v0.17.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI= golang.org/x/sync v0.18.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI= -golang.org/x/sync v0.19.0 h1:vV+1eWNmZ5geRlYjzm2adRgW2/mcpevXNg50YZtPCE4= -golang.org/x/sync v0.19.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI= golang.org/x/sys v0.21.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA= golang.org/x/sys v0.32.0/go.mod h1:BJP2sWEmIv4KK5OTEluFJCKSidICx8ciO85XgH3Ak8k= golang.org/x/sys v0.35.0/go.mod h1:BJP2sWEmIv4KK5OTEluFJCKSidICx8ciO85XgH3Ak8k= @@ -111,10 +111,10 @@ golang.org/x/term v0.40.0/go.mod h1:w2P8uVp06p2iyKKuvXIm7N/y0UCRt3UfJTfZ7oOpglM= golang.org/x/text v0.28.0/go.mod h1:U8nCwOR8jO/marOQ0QbDiOngZVEBB7MAiitBuMjXiNU= golang.org/x/tools v0.22.0/go.mod h1:aCwcsjqvq7Yqt6TNyX7QMU2enbQ/Gt0bo6krSeEri+c= golang.org/x/tools v0.37.0/go.mod h1:MBN5QPQtLMHVdvsbtarmTNukZDdgwdwlO5qGacAzF0w= -golang.org/x/tools v0.41.0/go.mod h1:XSY6eDqxVNiYgezAVqqCeihT4j1U2CCsqvH3WhQpnlg= golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543 h1:E7g+9GITq07hpfrRu66IVDexMakfv52eLZ2CXBWiKr4= golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0= golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0= +golang.org/x/xerrors v0.0.0-20231012003039-104605ab7028/go.mod h1:NDW/Ps6MPRej6fsCIbMTohpP40sJ/P/vI1MoTEGwX90= google.golang.org/appengine v1.6.7 h1:FZR1q0exgwxzPzp/aF+VccGrSfxfPpkBqjIIEq3ru6c= google.golang.org/appengine v1.6.7/go.mod h1:8WjMMxjGQR8xUklV/ARdw2HLXBOI7O7uCIDZVag1xfc= google.golang.org/protobuf v1.26.0-rc.1/go.mod h1:jlhhOSvTdKEhbULTjvd4ARK9grFBp09yW+WbY/TyQbw= @@ -125,8 +125,11 @@ gopkg.in/ini.v1 v1.67.0 h1:Dgnx+6+nfE+IfzjUEISNeydPJh9AXNNsWbGP9KzCsOA= gopkg.in/ini.v1 v1.67.0/go.mod h1:pNLf8WUiyNEtQjuu5G5vTm06TEv9tsIgeAvK8hOrP4k= gopkg.in/yaml.v2 v2.2.2/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI= lukechampine.com/uint128 v1.2.0/go.mod h1:c4eWIwlEGaxC/+H1VguhU4PHXNWDCDMUlWdIWl2j1gk= +lukechampine.com/uint128 v1.3.0/go.mod h1:c4eWIwlEGaxC/+H1VguhU4PHXNWDCDMUlWdIWl2j1gk= modernc.org/cc/v3 v3.40.0/go.mod h1:/bTg4dnWkSXowUO6ssQKnOV0yMVxDYNIsIrzqTFDGH0= +modernc.org/cc/v3 v3.41.0/go.mod h1:Ni4zjJYJ04CDOhG7dn640WGfwBzfE0ecX8TyMB0Fv0Y= modernc.org/ccgo/v3 v3.16.13/go.mod h1:2Quk+5YgpImhPjv2Qsob1DnZ/4som1lJTodubIcoUkY= +modernc.org/ccgo/v3 v3.16.15/go.mod h1:yT7B+/E2m43tmMOT51GMoM98/MtHIcQQSleGnddkUNI= modernc.org/httpfs v1.0.6/go.mod h1:7dosgurJGp0sPaRanU53W4xZYKh14wfzX420oZADeHM= modernc.org/opt v0.1.3/go.mod h1:WdSiB5evDcignE70guQKxYUl14mgWtbClRi5wmkkTX0= modernc.org/strutil v1.1.3/go.mod h1:MEHNA7PdEnEwLvspRMtWTNnp2nnyvMfkimT1NKNAGbw= From 2cd19d896495f6fb71835614866290cb7632e9a8 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 02:46:41 +0000 Subject: [PATCH 08/38] fix(uptime): implement SyncAndCheckForHost and cleanup stale failure counts; add tests for concurrency and feature flag handling --- .../api/handlers/proxy_host_handler.go | 14 +- backend/internal/api/routes/routes.go | 14 +- .../internal/models/notification_provider.go | 2 +- backend/internal/services/uptime_service.go | 117 +++ .../services/uptime_service_pr1_test.go | 421 +++++++++ docs/plans/current_spec.md | 806 +++++++++++++----- 6 files changed, 1153 insertions(+), 221 deletions(-) create mode 100644 backend/internal/services/uptime_service_pr1_test.go diff --git a/backend/internal/api/handlers/proxy_host_handler.go b/backend/internal/api/handlers/proxy_host_handler.go index 31750731..5ab90db2 100644 --- a/backend/internal/api/handlers/proxy_host_handler.go +++ b/backend/internal/api/handlers/proxy_host_handler.go @@ -413,6 +413,11 @@ func (h *ProxyHostHandler) Create(c *gin.Context) { ) } + // Trigger immediate uptime monitor creation + health check (non-blocking) + if h.uptimeService != nil { + go h.uptimeService.SyncAndCheckForHost(host.ID) + } + // Generate advisory warnings for private/Docker IPs warnings := generateForwardHostWarnings(host.ForwardHost) @@ -645,11 +650,10 @@ func (h *ProxyHostHandler) Delete(c *gin.Context) { return } - // check if we should also delete associated uptime monitors (query param: delete_uptime=true) - deleteUptime := c.DefaultQuery("delete_uptime", "false") == "true" - - if deleteUptime && h.uptimeService != nil { - // Find all monitors referencing this proxy host and delete each + // Always clean up associated uptime monitors when deleting a proxy host. + // The query param delete_uptime=true is kept for backward compatibility but + // cleanup now runs unconditionally to prevent orphaned monitors. + if h.uptimeService != nil { var monitors []models.UptimeMonitor if err := h.uptimeService.DB.Where("proxy_host_id = ?", host.ID).Find(&monitors).Error; err == nil { for _, m := range monitors { diff --git a/backend/internal/api/routes/routes.go b/backend/internal/api/routes/routes.go index cbd9881d..2382c575 100644 --- a/backend/internal/api/routes/routes.go +++ b/backend/internal/api/routes/routes.go @@ -410,9 +410,10 @@ func RegisterWithDeps(router *gin.Engine, db *gorm.DB, cfg config.Config, caddyM dockerHandler := handlers.NewDockerHandler(dockerService, remoteServerService) dockerHandler.RegisterRoutes(protected) - // Uptime Service - uptimeSvc := services.NewUptimeService(db, notificationService) - uptimeHandler := handlers.NewUptimeHandler(uptimeSvc) + // Uptime Service — reuse the single uptimeService instance (defined above) + // to share in-memory state (mutexes, notification batching) between + // background checker, ProxyHostHandler, and API handlers. + uptimeHandler := handlers.NewUptimeHandler(uptimeService) protected.GET("/uptime/monitors", uptimeHandler.List) protected.POST("/uptime/monitors", uptimeHandler.Create) protected.GET("/uptime/monitors/:id/history", uptimeHandler.GetHistory) @@ -464,9 +465,16 @@ func RegisterWithDeps(router *gin.Engine, db *gorm.DB, cfg config.Config, caddyM } if enabled { + // Clean up stale failure counts from historical bugs before first sync + if err := uptimeService.CleanupStaleFailureCounts(); err != nil { + logger.Log().WithError(err).Warn("Failed to cleanup stale failure counts") + } + if err := uptimeService.SyncMonitors(); err != nil { logger.Log().WithError(err).Error("Failed to sync monitors") } + // Run initial check immediately after sync to avoid the 90s blind window + uptimeService.CheckAll() } ticker := time.NewTicker(1 * time.Minute) diff --git a/backend/internal/models/notification_provider.go b/backend/internal/models/notification_provider.go index d31cf5c2..9d6427ec 100644 --- a/backend/internal/models/notification_provider.go +++ b/backend/internal/models/notification_provider.go @@ -14,7 +14,7 @@ type NotificationProvider struct { Type string `json:"type" gorm:"index"` // discord (only supported type in current rollout) URL string `json:"url"` // Discord webhook URL (HTTPS format required) Token string `json:"-"` // Auth token for providers (e.g., Gotify) - never exposed in API - HasToken bool `json:"has_token" gorm:"-"` // Computed: indicates whether a token is set (never exposes raw value) + HasToken bool `json:"has_token" gorm:"-"` // Computed: indicates whether a token is set (never exposes raw value) Engine string `json:"engine,omitempty" gorm:"index"` // notify_v1 (notify-only runtime) Config string `json:"config"` // JSON payload template for custom webhooks ServiceConfig string `json:"service_config,omitempty" gorm:"type:text"` // JSON blob for typed service config diff --git a/backend/internal/services/uptime_service.go b/backend/internal/services/uptime_service.go index 6da26b83..33030392 100644 --- a/backend/internal/services/uptime_service.go +++ b/backend/internal/services/uptime_service.go @@ -1184,3 +1184,120 @@ func (s *UptimeService) DeleteMonitor(id string) error { return nil } + +// SyncAndCheckForHost creates a monitor for the given proxy host (if one +// doesn't already exist) and immediately triggers a health check in a +// background goroutine. It is safe to call from any goroutine. +// +// Designed to be called as `go svc.SyncAndCheckForHost(hostID)` so it +// does not block the API response. +func (s *UptimeService) SyncAndCheckForHost(hostID uint) { + // Check feature flag — bail if uptime is disabled + var setting models.Setting + if err := s.DB.Where("key = ?", "feature.uptime.enabled").First(&setting).Error; err == nil { + if setting.Value != "true" { + return + } + } + + // Per-host lock prevents duplicate monitors when multiple goroutines + // call SyncAndCheckForHost for the same hostID concurrently. + hostKey := fmt.Sprintf("proxy-%d", hostID) + s.hostMutexLock.Lock() + if s.hostMutexes[hostKey] == nil { + s.hostMutexes[hostKey] = &sync.Mutex{} + } + mu := s.hostMutexes[hostKey] + s.hostMutexLock.Unlock() + + mu.Lock() + defer mu.Unlock() + + // Look up the proxy host; it may have been deleted between the API + // response and this goroutine executing. + var host models.ProxyHost + if err := s.DB.Where("id = ?", hostID).First(&host).Error; err != nil { + logger.Log().WithField("host_id", hostID).Debug("SyncAndCheckForHost: proxy host not found (may have been deleted)") + return + } + + // Ensure a monitor exists for this host + var monitor models.UptimeMonitor + err := s.DB.Where("proxy_host_id = ?", host.ID).First(&monitor).Error + if errors.Is(err, gorm.ErrRecordNotFound) { + domains := strings.Split(host.DomainNames, ",") + firstDomain := "" + if len(domains) > 0 { + firstDomain = strings.TrimSpace(domains[0]) + } + + scheme := "http" + if host.SSLForced { + scheme = "https" + } + publicURL := fmt.Sprintf("%s://%s", scheme, firstDomain) + upstreamHost := host.ForwardHost + + name := host.Name + if name == "" { + name = firstDomain + } + + uptimeHostID := s.ensureUptimeHost(upstreamHost, name) + + monitor = models.UptimeMonitor{ + ProxyHostID: &host.ID, + UptimeHostID: &uptimeHostID, + Name: name, + Type: "http", + URL: publicURL, + UpstreamHost: upstreamHost, + Interval: 60, + Enabled: true, + Status: "pending", + } + if createErr := s.DB.Create(&monitor).Error; createErr != nil { + logger.Log().WithError(createErr).WithField("host_id", host.ID).Error("SyncAndCheckForHost: failed to create monitor") + return + } + } else if err != nil { + logger.Log().WithError(err).WithField("host_id", host.ID).Error("SyncAndCheckForHost: failed to query monitor") + return + } + + // Run health check immediately + s.checkMonitor(monitor) +} + +// CleanupStaleFailureCounts resets monitors that are stuck in "down" status +// with elevated failure counts from historical bugs (e.g., port mismatch era). +// Only resets monitors with no recent successful heartbeat in the last 24 hours. +func (s *UptimeService) CleanupStaleFailureCounts() error { + result := s.DB.Exec(` + UPDATE uptime_monitors SET failure_count = 0, status = 'pending' + WHERE status = 'down' + AND failure_count > 5 + AND id NOT IN ( + SELECT DISTINCT monitor_id FROM uptime_heartbeats + WHERE status = 'up' AND created_at > datetime('now', '-24 hours') + ) + `) + if result.Error != nil { + return fmt.Errorf("cleanup stale failure counts: %w", result.Error) + } + + if result.RowsAffected > 0 { + logger.Log().WithField("reset_count", result.RowsAffected).Info("Reset stale monitor failure counts") + } + + hostResult := s.DB.Exec(`UPDATE uptime_hosts SET failure_count = 0, status = 'pending' WHERE status = 'down'`) + if hostResult.Error != nil { + return fmt.Errorf("cleanup stale host failure counts: %w", hostResult.Error) + } + + if hostResult.RowsAffected > 0 { + logger.Log().WithField("reset_count", hostResult.RowsAffected).Info("Reset stale host failure counts") + } + + return nil +} diff --git a/backend/internal/services/uptime_service_pr1_test.go b/backend/internal/services/uptime_service_pr1_test.go new file mode 100644 index 00000000..7c6b425e --- /dev/null +++ b/backend/internal/services/uptime_service_pr1_test.go @@ -0,0 +1,421 @@ +package services + +import ( + "fmt" + "os" + "path/filepath" + "sync" + "testing" + "time" + + "github.com/google/uuid" + "github.com/stretchr/testify/assert" + "github.com/stretchr/testify/require" + "gorm.io/driver/sqlite" + "gorm.io/gorm" + + "github.com/Wikid82/charon/backend/internal/models" +) + +// setupPR1TestDB creates an in-memory SQLite database with all models needed +// for PR-1 uptime bug fix tests. +func setupPR1TestDB(t *testing.T) *gorm.DB { + t.Helper() + dir := t.TempDir() + dbPath := filepath.Join(dir, "pr1test.db") + dsn := dbPath + "?_journal_mode=WAL&_busy_timeout=5000" + db, err := gorm.Open(sqlite.Open(dsn), &gorm.Config{}) + require.NoError(t, err) + require.NoError(t, db.AutoMigrate( + &models.UptimeMonitor{}, + &models.UptimeHeartbeat{}, + &models.UptimeHost{}, + &models.ProxyHost{}, + &models.Setting{}, + )) + + t.Cleanup(func() { + sqlDB, _ := db.DB() + if sqlDB != nil { + _ = sqlDB.Close() + } + }) + + return db +} + +// enableUptimeFeature sets the feature.uptime.enabled setting to "true". +func enableUptimeFeature(t *testing.T, db *gorm.DB) { + t.Helper() + require.NoError(t, db.Create(&models.Setting{ + Key: "feature.uptime.enabled", + Value: "true", + Type: "bool", + Category: "feature", + }).Error) +} + +// createTestProxyHost creates a minimal proxy host for testing. +func createTestProxyHost(t *testing.T, db *gorm.DB, name, domain, forwardHost string) models.ProxyHost { + t.Helper() + host := models.ProxyHost{ + UUID: uuid.New().String(), + Name: name, + DomainNames: domain, + ForwardScheme: "http", + ForwardHost: forwardHost, + ForwardPort: 80, + Enabled: true, + } + require.NoError(t, db.Create(&host).Error) + return host +} + +// --- Fix 1: Singleton UptimeService --- + +func TestSingletonUptimeService_SharedState(t *testing.T) { + db := setupPR1TestDB(t) + svc := NewUptimeService(db, nil) + + // Verify both pendingNotifications and hostMutexes are the same instance + // by writing to the maps from the shared reference. + svc.pendingNotifications["test-key"] = &pendingHostNotification{} + assert.Contains(t, svc.pendingNotifications, "test-key", + "pendingNotifications should be shared on the same instance") + + // A second reference to the same service should see the same map state. + svc2 := svc // simulate routes.go passing the same pointer + assert.Contains(t, svc2.pendingNotifications, "test-key", + "second reference must share the same pendingNotifications map") +} + +// --- Fix 2: SyncAndCheckForHost --- + +func TestSyncAndCheckForHost_CreatesMonitorAndHeartbeat(t *testing.T) { + db := setupPR1TestDB(t) + enableUptimeFeature(t, db) + svc := NewUptimeService(db, nil) + + host := createTestProxyHost(t, db, "test-host", "example.com", "192.168.1.100") + + // Execute synchronously (normally called as goroutine) + svc.SyncAndCheckForHost(host.ID) + + // Verify monitor was created + var monitor models.UptimeMonitor + err := db.Where("proxy_host_id = ?", host.ID).First(&monitor).Error + require.NoError(t, err, "monitor should be created for the proxy host") + assert.Equal(t, "http://example.com", monitor.URL) + assert.Equal(t, "192.168.1.100", monitor.UpstreamHost) + assert.Contains(t, []string{"up", "down", "pending"}, monitor.Status, "status should be set by checkMonitor") + + // Verify at least one heartbeat was created (from the immediate check) + var hbCount int64 + db.Model(&models.UptimeHeartbeat{}).Where("monitor_id = ?", monitor.ID).Count(&hbCount) + assert.Greater(t, hbCount, int64(0), "at least one heartbeat should exist after SyncAndCheckForHost") +} + +func TestSyncAndCheckForHost_SSLForcedUsesHTTPS(t *testing.T) { + db := setupPR1TestDB(t) + enableUptimeFeature(t, db) + svc := NewUptimeService(db, nil) + + host := models.ProxyHost{ + UUID: uuid.New().String(), + Name: "ssl-host", + DomainNames: "secure.example.com", + ForwardScheme: "https", + ForwardHost: "192.168.1.200", + ForwardPort: 443, + SSLForced: true, + Enabled: true, + } + require.NoError(t, db.Create(&host).Error) + + svc.SyncAndCheckForHost(host.ID) + + var monitor models.UptimeMonitor + require.NoError(t, db.Where("proxy_host_id = ?", host.ID).First(&monitor).Error) + assert.Equal(t, "https://secure.example.com", monitor.URL) +} + +func TestSyncAndCheckForHost_DeletedHostNoPanic(t *testing.T) { + db := setupPR1TestDB(t) + enableUptimeFeature(t, db) + svc := NewUptimeService(db, nil) + + // Call with a host ID that doesn't exist — should log and return, not panic + assert.NotPanics(t, func() { + svc.SyncAndCheckForHost(99999) + }) + + // No monitor should be created + var count int64 + db.Model(&models.UptimeMonitor{}).Count(&count) + assert.Equal(t, int64(0), count) +} + +func TestSyncAndCheckForHost_ExistingMonitorSkipsCreate(t *testing.T) { + db := setupPR1TestDB(t) + enableUptimeFeature(t, db) + svc := NewUptimeService(db, nil) + + host := createTestProxyHost(t, db, "existing-mon", "existing.com", "10.0.0.1") + + // Pre-create a monitor + existingMonitor := models.UptimeMonitor{ + ID: uuid.New().String(), + ProxyHostID: &host.ID, + Name: "pre-existing", + Type: "http", + URL: "http://existing.com", + Interval: 60, + Enabled: true, + Status: "up", + } + require.NoError(t, db.Create(&existingMonitor).Error) + + svc.SyncAndCheckForHost(host.ID) + + // Should still be exactly 1 monitor + var count int64 + db.Model(&models.UptimeMonitor{}).Where("proxy_host_id = ?", host.ID).Count(&count) + assert.Equal(t, int64(1), count, "should not create a duplicate monitor") +} + +// --- Fix 2 continued: Feature flag test --- + +func TestSyncAndCheckForHost_DisabledFeatureNoop(t *testing.T) { + db := setupPR1TestDB(t) + // Explicitly set feature to disabled + require.NoError(t, db.Create(&models.Setting{ + Key: "feature.uptime.enabled", + Value: "false", + Type: "bool", + Category: "feature", + }).Error) + svc := NewUptimeService(db, nil) + + host := createTestProxyHost(t, db, "disabled-host", "disabled.com", "10.0.0.2") + + svc.SyncAndCheckForHost(host.ID) + + // No monitor should be created when feature is disabled + var count int64 + db.Model(&models.UptimeMonitor{}).Where("proxy_host_id = ?", host.ID).Count(&count) + assert.Equal(t, int64(0), count, "no monitor should be created when feature is disabled") +} + +func TestSyncAndCheckForHost_MissingSetting_StillCreates(t *testing.T) { + db := setupPR1TestDB(t) + // No setting at all — the method should proceed (default: enabled behavior) + svc := NewUptimeService(db, nil) + + host := createTestProxyHost(t, db, "no-setting", "nosetting.com", "10.0.0.3") + + svc.SyncAndCheckForHost(host.ID) + + var count int64 + db.Model(&models.UptimeMonitor{}).Where("proxy_host_id = ?", host.ID).Count(&count) + assert.Greater(t, count, int64(0), "monitor should be created when setting is missing (default: enabled)") +} + +// --- Fix 4: CleanupStaleFailureCounts --- + +func TestCleanupStaleFailureCounts_ResetsStuckMonitors(t *testing.T) { + db := setupPR1TestDB(t) + svc := NewUptimeService(db, nil) + + // Create a "stuck" monitor: down, failure_count > 5, no recent UP heartbeat + stuckMonitor := models.UptimeMonitor{ + ID: uuid.New().String(), + Name: "stuck-monitor", + Type: "http", + URL: "http://stuck.example.com", + Interval: 60, + Enabled: true, + Status: "down", + FailureCount: 10, + } + require.NoError(t, db.Create(&stuckMonitor).Error) + + err := svc.CleanupStaleFailureCounts() + require.NoError(t, err) + + // Verify the monitor was reset + var m models.UptimeMonitor + require.NoError(t, db.First(&m, "id = ?", stuckMonitor.ID).Error) + assert.Equal(t, 0, m.FailureCount, "failure_count should be reset to 0") + assert.Equal(t, "pending", m.Status, "status should be reset to pending") +} + +func TestCleanupStaleFailureCounts_SkipsMonitorsWithRecentUpHeartbeat(t *testing.T) { + db := setupPR1TestDB(t) + svc := NewUptimeService(db, nil) + + // Create a monitor that is "down" with high failure_count BUT has a recent UP heartbeat + healthyMonitor := models.UptimeMonitor{ + ID: uuid.New().String(), + Name: "healthy-monitor", + Type: "http", + URL: "http://healthy.example.com", + Interval: 60, + Enabled: true, + Status: "down", + FailureCount: 10, + } + require.NoError(t, db.Create(&healthyMonitor).Error) + + // Add a recent UP heartbeat + hb := models.UptimeHeartbeat{ + MonitorID: healthyMonitor.ID, + Status: "up", + Latency: 50, + CreatedAt: time.Now().Add(-1 * time.Hour), // 1 hour ago — within 24h window + } + require.NoError(t, db.Create(&hb).Error) + + err := svc.CleanupStaleFailureCounts() + require.NoError(t, err) + + // Monitor should NOT be reset because it has a recent UP heartbeat + var m models.UptimeMonitor + require.NoError(t, db.First(&m, "id = ?", healthyMonitor.ID).Error) + assert.Equal(t, 10, m.FailureCount, "failure_count should NOT be reset since there's a recent UP heartbeat") + assert.Equal(t, "down", m.Status, "status should remain down") +} + +func TestCleanupStaleFailureCounts_SkipsLowFailureCount(t *testing.T) { + db := setupPR1TestDB(t) + svc := NewUptimeService(db, nil) + + // Monitor with failure_count <= 5 — should not be touched + monitor := models.UptimeMonitor{ + ID: uuid.New().String(), + Name: "low-failure-monitor", + Type: "http", + URL: "http://low.example.com", + Interval: 60, + Enabled: true, + Status: "down", + FailureCount: 3, + } + require.NoError(t, db.Create(&monitor).Error) + + err := svc.CleanupStaleFailureCounts() + require.NoError(t, err) + + var m models.UptimeMonitor + require.NoError(t, db.First(&m, "id = ?", monitor.ID).Error) + assert.Equal(t, 3, m.FailureCount, "low failure_count should not be reset") + assert.Equal(t, "down", m.Status) +} + +func TestCleanupStaleFailureCounts_ResetsStaleHosts(t *testing.T) { + db := setupPR1TestDB(t) + svc := NewUptimeService(db, nil) + + // Create a "stuck" host + host := models.UptimeHost{ + ID: uuid.New().String(), + Host: "stuck-host.local", + Name: "stuck-host", + Status: "down", + FailureCount: 10, + } + require.NoError(t, db.Create(&host).Error) + + err := svc.CleanupStaleFailureCounts() + require.NoError(t, err) + + var h models.UptimeHost + require.NoError(t, db.First(&h, "id = ?", host.ID).Error) + assert.Equal(t, 0, h.FailureCount) + assert.Equal(t, "pending", h.Status) +} + +// setupPR1ConcurrentDB creates a file-based SQLite database with WAL mode and +// busy_timeout to handle concurrent writes without "database table is locked". +func setupPR1ConcurrentDB(t *testing.T) *gorm.DB { + t.Helper() + dir := t.TempDir() + dbPath := filepath.Join(dir, "test.db") + dsn := dbPath + "?_journal_mode=WAL&_busy_timeout=5000" + db, err := gorm.Open(sqlite.Open(dsn), &gorm.Config{}) + require.NoError(t, err) + require.NoError(t, db.AutoMigrate( + &models.UptimeMonitor{}, + &models.UptimeHeartbeat{}, + &models.UptimeHost{}, + &models.ProxyHost{}, + &models.Setting{}, + )) + + t.Cleanup(func() { + sqlDB, _ := db.DB() + if sqlDB != nil { + _ = sqlDB.Close() + } + _ = os.Remove(dbPath) + }) + + return db +} + +// --- Concurrent access tests --- + +func TestSyncAndCheckForHost_ConcurrentCreates_NoDuplicates(t *testing.T) { + db := setupPR1ConcurrentDB(t) + enableUptimeFeature(t, db) + svc := NewUptimeService(db, nil) + + // Create multiple proxy hosts with unique domains + hosts := make([]models.ProxyHost, 5) + for i := range hosts { + hosts[i] = createTestProxyHost(t, db, + fmt.Sprintf("concurrent-host-%d", i), + fmt.Sprintf("concurrent-%d.com", i), + fmt.Sprintf("10.0.0.%d", 100+i), + ) + } + + var wg sync.WaitGroup + for _, h := range hosts { + wg.Add(1) + go func(hostID uint) { + defer wg.Done() + svc.SyncAndCheckForHost(hostID) + }(h.ID) + } + wg.Wait() + + // Each host should have exactly 1 monitor + for _, h := range hosts { + var count int64 + db.Model(&models.UptimeMonitor{}).Where("proxy_host_id = ?", h.ID).Count(&count) + assert.Equal(t, int64(1), count, "each proxy host should have exactly 1 monitor") + } +} + +func TestSyncAndCheckForHost_ConcurrentSameHost_NoDuplicates(t *testing.T) { + db := setupPR1ConcurrentDB(t) + enableUptimeFeature(t, db) + svc := NewUptimeService(db, nil) + + host := createTestProxyHost(t, db, "race-host", "race.com", "10.0.0.200") + + var wg sync.WaitGroup + for i := 0; i < 10; i++ { + wg.Add(1) + go func() { + defer wg.Done() + svc.SyncAndCheckForHost(host.ID) + }() + } + wg.Wait() + + // Should still be exactly 1 monitor even after 10 concurrent calls + var count int64 + db.Model(&models.UptimeMonitor{}).Where("proxy_host_id = ?", host.ID).Count(&count) + assert.Equal(t, int64(1), count, "concurrent SyncAndCheckForHost should not create duplicates") +} diff --git a/docs/plans/current_spec.md b/docs/plans/current_spec.md index 81fc1c46..40be9842 100644 --- a/docs/plans/current_spec.md +++ b/docs/plans/current_spec.md @@ -1,270 +1,652 @@ -# ACL + Security Headers Hotfix Plan (Proxy Host Create/Edit) +# Uptime Monitoring Bug Triage & Fix Plan ## 1. Introduction ### Overview -Hotfix request: Proxy Host form dropdown selections for Access Control List (ACL) and Security Headers are not being applied/persisted for new or edited hosts. -Reported behavior: -1. Existing hosts with previously assigned ACL/Security Header profile retain old values. -2. Users cannot reliably remove or change those values in UI. -3. Newly created hosts cannot reliably apply ACL/Security Header profile. +Uptime Monitoring in Charon uses a two-level check system: host-level TCP pre-checks followed by per-monitor HTTP/TCP checks. Newly added proxy hosts (specifically Wizarr and Charon itself) display as "DOWN" in the UI even though the underlying services are fully accessible. Manual refresh via the health check button on the Uptime page correctly shows "UP", but the automated background checker fails to produce the same result. -### Objective -Deliver an urgent but correct root-cause fix across frontend binding and backend persistence flow, with minimum user interruption and full validation gates. +### Objectives -## 2. Research Findings (Current Architecture + Touchpoints) +1. Eliminate false "DOWN" status for newly added proxy hosts +2. Ensure the background checker produces consistent results with manual health checks +3. Improve the initial monitor lifecycle (creation → first check → display) +4. Address the dual `UptimeService` instance functional inconsistency +5. Evaluate whether a "custom health endpoint URL" feature is warranted -### Frontend Entry Points -1. `frontend/src/pages/ProxyHosts.tsx` - - `handleSubmit(data)` calls `updateHost(editingHost.uuid, data)` or `createHost(data)`. - - Renders `ProxyHostForm` modal for create/edit flows. -2. `frontend/src/components/ProxyHostForm.tsx` - - Local form state initializes `access_list_id` and `security_header_profile_id`. - - ACL control uses `AccessListSelector`. - - Security Headers control uses `Select` with `security_header_profile_id` mapping. - - Submission path: `handleSubmit` -> `onSubmit(payloadWithoutUptime)`. -3. `frontend/src/components/AccessListSelector.tsx` - - Converts select values between `string` and `number | null`. +### Scope -### Frontend API/Hooks -1. `frontend/src/hooks/useProxyHosts.ts` - - `createHost` -> `createProxyHost`. - - `updateHost` -> `updateProxyHost`. -2. `frontend/src/api/proxyHosts.ts` - - `createProxyHost(host: Partial)` -> `POST /api/v1/proxy-hosts`. - - `updateProxyHost(uuid, host)` -> `PUT /api/v1/proxy-hosts/:uuid`. - - Contract fields: `access_list_id`, `security_header_profile_id`. +- **Backend**: `backend/internal/services/uptime_service.go`, `backend/internal/api/routes/routes.go`, `backend/internal/api/handlers/proxy_host_handler.go` +- **Frontend**: `frontend/src/pages/Uptime.tsx`, `frontend/src/api/uptime.ts` +- **Models**: `backend/internal/models/uptime.go`, `backend/internal/models/uptime_host.go` +- **Tests**: `backend/internal/services/uptime_service_test.go` (1519 LOC), `uptime_service_unit_test.go` (257 LOC), `uptime_service_race_test.go` (402 LOC), `tests/monitoring/uptime-monitoring.spec.ts` (E2E) -### Backend Entry/Transformation/Persistence -1. Route registration - - `backend/internal/api/routes/routes.go`: `proxyHostHandler.RegisterRoutes(protected)`. -2. Handler - - `backend/internal/api/handlers/proxy_host_handler.go` - - `Create(c)` uses `ShouldBindJSON(&models.ProxyHost{})`. - - `Update(c)` uses `map[string]any` partial update parsing. - - Target fields: - - `payload["access_list_id"]` -> `parseNullableUintField` -> `host.AccessListID` - - `payload["security_header_profile_id"]` -> typed conversion -> `host.SecurityHeaderProfileID` -3. Service - - `backend/internal/services/proxyhost_service.go` - - `Create(host)` validates + `db.Create(host)`. - - `Update(host)` validates + `db.Model(...).Select("*").Updates(host)`. -4. Model - - `backend/internal/models/proxy_host.go` - - `AccessListID *uint \`json:"access_list_id"\`` - - `SecurityHeaderProfileID *uint \`json:"security_header_profile_id"\`` +--- -### Existing Tests Relevant to Incident -1. Frontend unit regression coverage already exists: - - `frontend/src/components/__tests__/ProxyHostForm-dropdown-changes.test.tsx` -2. E2E regression spec exists: - - `tests/security-enforcement/acl-dropdown-regression.spec.ts` -3. Backend update and security-header tests exist: - - `backend/internal/api/handlers/proxy_host_handler_update_test.go` - - `backend/internal/api/handlers/proxy_host_handler_security_headers_test.go` +## 2. Research Findings -## 3. Root-Cause-First Trace +### 2.1 Root Cause #1: Port Mismatch in Host-Level TCP Check (FIXED) -### Trace Model (Mandatory) -1. Entry Point: - - UI dropdown interactions in `ProxyHostForm` and `AccessListSelector`. -2. Transformation: - - Form state conversion (`string` <-> `number | null`) and payload construction in `ProxyHostForm`. - - API serialization via `frontend/src/api/proxyHosts.ts`. -3. Persistence: - - Backend `Update` parser (`proxy_host_handler.go`) and `ProxyHostService.Update` persistence. -4. Exit Point: - - Response body consumed by React Query invalidation/refetch in `useProxyHosts`. - - UI reflects updated values in table/form. +**Status**: Fixed in commit `209b2fc8`, refactored in `bfc19ef3`. -### Most Likely Failure Zones -1. Frontend select binding/conversion drift (top candidate) - - Shared symptom across ACL and Security Headers points to form/select layer. - - Candidate files: - - `frontend/src/components/ProxyHostForm.tsx` - - `frontend/src/components/AccessListSelector.tsx` - - `frontend/src/components/ui/Select.tsx` -2. Payload mutation or stale form object behavior - - Ensure payload carries updated `access_list_id` / `security_header_profile_id` values at submit time. -3. Backend partial-update parser edge behavior - - Ensure `nil`, numeric string, and number conversions are consistent between ACL and security header profile paths. +The `checkHost()` function extracted the port from the monitor's public URL (e.g., 443 for HTTPS) instead of using `ProxyHost.ForwardPort` (e.g., 5690 for Wizarr). This caused TCP checks to fail, marking the host as `down`, which then skipped individual HTTP monitor checks. -### Investigation Decision -Root-cause verification will be instrumented through failing-first Playwright scenario and targeted handler tests before applying code changes. +**Fix applied**: Added `Preload("ProxyHost")` and prioritized `monitor.ProxyHost.ForwardPort` over `extractPort(monitor.URL)`. -## 4. EARS Requirements +**Evidence**: Archived in `docs/plans/archive/uptime_monitoring_diagnosis.md` and `docs/implementation/uptime_monitoring_port_fix_COMPLETE.md`. -1. WHEN a user selects an ACL in the Proxy Host create/edit form, THE SYSTEM SHALL persist `access_list_id` and return it in API response. -2. WHEN a user changes ACL from one value to another, THE SYSTEM SHALL replace prior `access_list_id` with the new value. -3. WHEN a user selects "No Access Control", THE SYSTEM SHALL persist `access_list_id = null`. -4. WHEN a user selects a Security Headers profile in the Proxy Host create/edit form, THE SYSTEM SHALL persist `security_header_profile_id` and return it in API response. -5. WHEN a user changes Security Headers profile from one value to another, THE SYSTEM SHALL replace prior `security_header_profile_id` with the new value. -6. WHEN a user selects "None" for Security Headers, THE SYSTEM SHALL persist `security_header_profile_id = null`. -7. IF dropdown interaction fails to update internal form state, THEN THE SYSTEM SHALL prevent stale values from being persisted. -8. WHILE updating Proxy Host settings, THE SYSTEM SHALL maintain existing behavior for unrelated fields and not regress certificate, DNS challenge, or uptime-linked updates. +**Remaining risk**: If this fix has not been deployed to production, this remains the primary cause. If deployed, residual elevated `failure_count` values in the DB may need to be reset. -Note: User-visible blocking error behavior is deferred unless required by confirmed root cause. +### 2.2 Root Cause #2: Dual UptimeService Instance (OPEN — Functional Inconsistency) -## 5. Technical Specification (Hotfix Scope) +**File**: `backend/internal/api/routes/routes.go` -### API Contract (No Breaking Change) -1. `POST /api/v1/proxy-hosts` - - Request fields include `access_list_id`, `security_header_profile_id` as nullable numeric fields. -2. `PUT /api/v1/proxy-hosts/:uuid` - - Partial payload accepts nullable updates for both fields. -3. Response must echo persisted values in snake_case: - - `access_list_id` - - `security_header_profile_id` +Two separate `UptimeService` instances are created: -### Data Model/DB -No schema migration expected. Existing nullable FK fields in `backend/internal/models/proxy_host.go` are sufficient. +| Instance | Line | Scope | +|----------|------|-------| +| `uptimeService` | 226 | Background ticker goroutine, `ProxyHostHandler`, `/system/uptime/check` endpoint | +| `uptimeSvc` | 414 | Uptime API handler routes (List, Create, Update, Delete, Check, Sync) | -### Targeted Code Areas for Fix -1. Frontend - - `frontend/src/components/ProxyHostForm.tsx` - - `frontend/src/components/AccessListSelector.tsx` - - `frontend/src/components/ui/Select.tsx` (only if click/select propagation issue confirmed) - - `frontend/src/api/proxyHosts.ts` (only if serialization issue confirmed) -2. Backend - - `backend/internal/api/handlers/proxy_host_handler.go` (only if parsing/persistence mismatch confirmed) - - `backend/internal/services/proxyhost_service.go` (only if update write path proves incorrect) +Both share the same `*gorm.DB` (so data consistency via DB is maintained), but each has **independent in-memory state**: -## 6. Edge Cases +- `pendingNotifications` map (notification batching) +- `hostMutexes` map (per-host mutex for concurrent writes) +- `batchWindow` timers -1. Edit host with existing ACL/profile and switch to another value. -2. Edit host with existing ACL/profile and clear to null. -3. Create new host with ACL/profile set before first save. -4. Submit with stringified numeric values (defensive compatibility). -5. Submit with null values for both fields simultaneously. -6. Missing/deleted profile or ACL IDs in backend (validation errors). -7. Multiple rapid dropdown changes before save (last selection wins). +**Impact**: This is a **functional inconsistency that can cause race conditions between ProxyHostHandler operations and Uptime API operations**. Specifically: -## 7. Risk Analysis +- `ProxyHostHandler.Create()` uses instance #1 (`uptimeService`) for `SyncAndCheckForHost` +- Uptime API queries (List, GetHistory) use instance #2 (`uptimeSvc`) +- In-memory state (host mutexes, pending notifications) is **invisible between instances** -### High Risk -1. Silent stale-state submission from form controls. -2. Regressing other Proxy Host settings due to broad payload mutation. +This creates a functional bug path because: -### Medium Risk -1. Partial-update parser divergence between ACL and security profile behavior. -2. UI select portal/z-index interaction causing non-deterministic click handling. +- When a user triggers a manual check via `POST /api/v1/uptime/monitors/:id/check`, the handler uses `uptimeSvc.CheckMonitor()`. If the monitor transitions to "down", the notification is queued in `uptimeSvc`'s `pendingNotifications` map. Meanwhile, the background checker uses `uptimeService`, which has a separate `pendingNotifications` map. +- Duplicate or missed notifications +- Independent failure debouncing state +- Mutex contention issues between the two instances -### Mitigations -1. Reproduce with Playwright first and capture exact failing action path. -2. Add/strengthen focused frontend tests around create/edit/clear flows. -3. Add/strengthen backend tests for nullable + conversion paths. -4. Keep hotfix minimal and avoid unrelated refactors. +While NOT the direct cause of the "DOWN" display bug, this is a functional inconsistency — not merely a code smell — that can produce observable bugs in notification delivery and state synchronization. -## 8. Implementation Plan (Urgent, Minimal Interruption) +### 2.3 Root Cause #3: No Immediate Monitor Creation on Proxy Host Create (OPEN) -### Phase 1: Reproduction + Guardrails (Playwright First) -1. Execute targeted E2E spec for dropdown flow and create/edit persistence behavior. -2. Capture exact failure step and confirm whether failure is click binding, payload value, or backend persistence. -3. Add/adjust failing-first test if current suite does not capture observed production regression. +> **Note — Create ↔ Update asymmetry**: `ProxyHostHandler.Update()` already calls `SyncMonitorForHost` (established pattern). The fix for `Create` should follow the same pattern for consistency. -### Phase 2: Frontend Fix -1. Patch select binding/state mapping for ACL and Security Headers in `ProxyHostForm`/`AccessListSelector`. -2. If needed, patch `ui/Select` interaction layering. -3. Ensure payload contains correct final `access_list_id` and `security_header_profile_id` values at submit. -4. Extend `ProxyHostForm` tests for create/edit/change/remove flows. +When a user creates a new proxy host: -### Phase 3: Backend Hardening (Conditional) -1. Only if frontend payload is correct but persistence is wrong: - - Backend fix MUST use field-scoped partial-update semantics for `access_list_id` and `security_header_profile_id` only (unless separately justified). - - Ensure write path persists null transitions reliably. -2. Add/adjust handler/service regression tests proving no unintended mutation of unrelated proxy host fields during these targeted updates. +1. The proxy host is saved to DB +2. **No uptime monitor is created** — there is no hook in `ProxyHostHandler.Create()` to trigger `SyncMonitors()` or create a monitor +3. `SyncMonitorForHost()` (called on proxy host update) only updates existing monitors — it does NOT create new ones +4. The background ticker must fire (up to 1 minute) for `SyncMonitors()` to create the monitor -### Phase 4: Integration + Regression -1. Run complete targeted Proxy Host UI flow tests. -2. Validate list refresh and modal reopen reflect persisted values. -3. Validate no regressions in bulk ACL / bulk security-header operations. +**Timeline for a new proxy host to show status**: -### Phase 5: Documentation + Handoff -1. Update changelog/release notes only for hotfix behavior. -2. Keep architecture docs unchanged unless root cause requires architectural note. -3. Handoff to Supervisor agent for review after plan approval and implementation. +- T+0s: Proxy host created via API +- T+0s to T+60s: No uptime monitor exists — Uptime page shows nothing for this host +- T+60s: Background ticker fires, `SyncMonitors()` creates monitor with `status: "pending"` +- T+60s: `CheckAll()` runs, attempts host check + individual check +- T+62s: If checks succeed, monitor `status: "up"` is saved to DB +- T+90s (worst case): Frontend polls monitors and picks up the update -## 9. Acceptance Criteria +This is a poor UX experience. Users expect to see their new host on the Uptime page immediately. -1. ACL dropdown selection persists on create and edit. -2. Security Headers dropdown selection persists on create and edit. -3. Clearing ACL persists `null` and is reflected after reload. -4. Clearing Security Headers persists `null` and is reflected after reload. -5. Existing hosts can change from one ACL/profile to another without stale value retention. -6. New hosts can apply ACL/profile at creation time. -7. No regressions in unrelated proxy host fields. -8. All validation gates in Section 11 pass. -9. API create response returns persisted `access_list_id` and `security_header_profile_id` matching submitted values (including `null`). -10. API update response returns persisted `access_list_id` and `security_header_profile_id` after `value->value`, `value->null`, and `null->value` transitions. -11. Backend persistence verification confirms unrelated proxy host fields remain unchanged for targeted updates. +### 2.4 Root Cause #4: "pending" Status Displayed as DOWN (OPEN) -## 10. PR Slicing Strategy +**File**: `frontend/src/pages/Uptime.tsx`, MonitorCard component -### Decision -Single PR (hotfix-first), with contingency split only if backend root cause is confirmed late. +```tsx +const isUp = latestBeat ? latestBeat.status === 'up' : monitor.status === 'up'; +``` -### Rationale -1. Incident impact is immediate user-facing and concentrated in one feature path. -2. Frontend + targeted backend/test changes are tightly coupled for verification. -3. Single PR minimizes release coordination and user interruption. +When a new monitor has `status: "pending"` and no heartbeat history: -### Contingency (Only if split becomes necessary) -1. PR-1: Frontend binding + tests - - Scope: `ProxyHostForm`, `AccessListSelector`, `ui/Select` (if required), related tests. - - Dependency: none. - - Acceptance: UI submit payload verified correct in unit + Playwright. -2. PR-2: Backend parser/persistence + tests (conditional) - - Scope: `proxy_host_handler.go`, `proxyhost_service.go`, handler/service tests. - - Dependency: PR-1 merged or rebased for aligned contract. - - Acceptance: API update/create persist both nullable IDs correctly. -3. PR-3: Regression hardening + docs - - Scope: extra regression coverage, release-note hotfix entry. - - Dependency: PR-1/PR-2. - - Acceptance: full DoD validation sequence passes. +- `latestBeat` = `null` (no history yet) +- Falls back to `monitor.status === 'up'` +- `"pending" === "up"` → `false` +- **Displayed with red DOWN styling** -## 11. Validation Plan (Mandatory Sequence) +The UI has no dedicated "pending" or "unknown" state. Between creation and first check, every monitor appears DOWN. -0. E2E environment prerequisite +### 2.5 Root Cause #5: No Initial CheckAll After Server Start Sync (OPEN) + +**File**: `backend/internal/api/routes/routes.go`, lines 455-490 + +The background goroutine flow on server start: + +1. Sleep 30 seconds +2. Call `SyncMonitors()` — creates monitors for all proxy hosts +3. **Does NOT call `CheckAll()`** +4. Start 1-minute ticker +5. First `CheckAll()` runs on first tick (~90 seconds after server start) + +This means after every server restart, all monitors sit in "pending" (displayed as DOWN) for up to 90 seconds. + +### 2.6 Concern #6: Self-Referencing Check (Charon Pinging Itself) + +If Charon has a proxy host pointing to itself (e.g., `charon.example.com` → `localhost:8080`): + +**TCP host check**: Connects to `localhost:8080` → succeeds (Gin server is running locally). + +**HTTP monitor check**: Sends GET to `https://charon.example.com` → requires DNS resolution from inside the Docker container. This may fail due to: + +- **Docker hairpin NAT**: Containers cannot reach their own published ports via the host's external IP by default +- **Split-horizon DNS**: The domain may resolve to a public IP that isn't routable from within the container +- **Caddy certificate validation**: The HTTP client might reject a self-signed or incorrectly configured cert + +When the user clicks manual refresh, the same `checkMonitor()` function runs with the same options (`WithAllowLocalhost()`, `WithMaxRedirects(0)`). If manual check succeeds but background check fails, the difference is likely **timing-dependent** — the alternating "up"/"down" pattern observed in the archived diagnosis (heartbeat records alternating between `up|HTTP 200` and `down|Host unreachable`) supports this hypothesis. + +### 2.7 Feature Gap: No Custom Health Endpoint URL + +The `UptimeMonitor` model has no `health_endpoint` or `custom_url` field. All monitors check the public root URL (`/`). This is problematic because: + +- Some services redirect root → `/login` → 302 → tracked inconsistently +- Services with dedicated health endpoints (`/health`, `/api/health`) provide more reliable status +- Self-referencing checks (Charon) could use `http://localhost:8080/api/v1/health` instead of routing through DNS/Caddy + +### 2.8 Existing Test Coverage + +| File | LOC | Focus | +|------|-----|-------| +| `uptime_service_test.go` | 1519 | Integration tests with SQLite DB | +| `uptime_service_unit_test.go` | 257 | Unit tests for service methods | +| `uptime_service_race_test.go` | 402 | Concurrency/race condition tests | +| `uptime_service_notification_test.go` | — | Notification batching tests | +| `uptime_handler_test.go` | — | Handler HTTP endpoint tests | +| `uptime_monitor_initial_state_test.go` | — | Initial state tests | +| `uptime-monitoring.spec.ts` | — | Playwright E2E (22 scenarios) | + +--- + +## 3. Technical Specifications + +### 3.1 Consolidate UptimeService Singleton + +**Current**: Two instances (`uptimeService` line 226, `uptimeSvc` line 414) in `routes.go`. + +**Target**: Single instance passed to both the background goroutine AND the API handlers. + +```go +// routes.go — BEFORE (two instances) +uptimeService := services.NewUptimeService(db, notificationService) // line 226 +uptimeSvc := services.NewUptimeService(db, notificationService) // line 414 + +// routes.go — AFTER (single instance) +uptimeService := services.NewUptimeService(db, notificationService) // line 226 +// line 414: reuse uptimeService for handler registration +uptimeHandler := handlers.NewUptimeHandler(uptimeService) +``` + +**Impact**: All in-memory state (mutexes, notification batching, pending notifications) is shared. The single instance must remain thread-safe (it already is — methods use `sync.Mutex`). + +### 3.2 Trigger Monitor Creation + Immediate Check on Proxy Host Create + +**File**: `backend/internal/api/handlers/proxy_host_handler.go` + +After successfully creating a proxy host, call `SyncMonitors()` (or a targeted sync) and trigger an immediate check: + +```go +// In Create handler, after host is saved: +if h.uptimeService != nil { + _ = h.uptimeService.SyncMonitors() + // Trigger immediate check for the new monitor + var monitor models.UptimeMonitor + if err := h.uptimeService.DB.Where("proxy_host_id = ?", host.ID).First(&monitor).Error; err == nil { + go h.uptimeService.CheckMonitor(monitor) + } +} +``` + +**Alternative (lighter-weight)**: Add a `SyncAndCheckForHost(hostID uint)` method that creates the monitor if needed and immediately checks it. + +### 3.3 Add "pending" UI State + +**File**: `frontend/src/pages/Uptime.tsx` + +Add dedicated handling for `"pending"` status: + +```tsx +const isPending = monitor.status === 'pending' && (!history || history.length === 0); +const isUp = latestBeat ? latestBeat.status === 'up' : monitor.status === 'up'; +const isPaused = monitor.enabled === false; +``` + +Visual treatment for pending state: + +- Yellow/gray pulsing indicator (distinct from DOWN red and UP green) +- Badge text: "CHECKING..." or "PENDING" +- Heartbeat bar: show empty placeholder bars with a spinner or pulse animation + +### 3.4 Run CheckAll After Initial SyncMonitors + +**File**: `backend/internal/api/routes/routes.go` + +```go +// AFTER initial sync +if enabled { + if err := uptimeService.SyncMonitors(); err != nil { + logger.Log().WithError(err).Error("Failed to sync monitors") + } + // Run initial check immediately + uptimeService.CheckAll() +} +``` + +### 3.5 Add Optional `check_url` Field to UptimeMonitor (Enhancement) + +**Model change** (`backend/internal/models/uptime.go`): + +```go +type UptimeMonitor struct { + // ... existing fields + CheckURL string `json:"check_url,omitempty" gorm:"default:null"` +} +``` + +**Service behavior** (`uptime_service.go` `checkMonitor()`): + +- If `monitor.CheckURL` is set and non-empty, use it instead of `monitor.URL` for the HTTP check +- This allows users to configure `/health` or `http://localhost:8080/api/v1/health` for self-referencing + +**Frontend**: Add an optional "Health Check URL" field in the edit monitor modal. + +**Auto-migration**: GORM handles adding the column. Existing monitors keep `CheckURL = ""` (uses default URL behavior). + +#### 3.5.1 SSRF Protection for CheckURL + +The `CheckURL` field accepts user-controlled URLs that the server will fetch. This requires layered SSRF defenses: + +**Write-time validation** (on Create/Update API): + +- Validate `CheckURL` before saving to DB +- **Scheme restriction**: Only `http://` and `https://` allowed. Block `file://`, `ftp://`, `gopher://`, and all other schemes +- **Max URL length**: 2048 characters +- Reject URLs that fail `url.Parse()` or have empty host components + +**Check-time validation** (before each HTTP request): + +- Re-validate the URL against the deny list before every check execution (defense-in-depth — the stored URL could have been valid at write time but conditions may change) +- **Localhost handling**: Allow loopback addresses (`127.0.0.1`, `::1`, `localhost`) since self-referencing checks are a valid use case. Block cloud metadata IPs: + - `169.254.169.254` (AWS/GCP/Azure instance metadata) + - `fd00::/8` (unique local addresses) + - `100.100.100.200` (Alibaba Cloud metadata) + - `169.254.0.0/16` link-local range (except loopback) +- **DNS rebinding protection**: Resolve the hostname at request time, pin the resolved IP, and validate the resolved IP against the deny list before establishing a connection. Use a custom `net.Dialer` or `http.Transport.DialContext` to enforce this +- **Redirect validation**: If `CheckURL` follows HTTP redirects (3xx), validate each redirect target URL against the same deny list (scheme, host, resolved IP). Use a `CheckRedirect` function on the `http.Client` to intercept and validate each hop + +**Implementation pattern**: + +```go +func validateCheckURL(rawURL string) error { + if len(rawURL) > 2048 { + return ErrURLTooLong + } + parsed, err := url.Parse(rawURL) + if err != nil { + return ErrInvalidURL + } + if parsed.Scheme != "http" && parsed.Scheme != "https" { + return ErrDisallowedScheme + } + if parsed.Host == "" { + return ErrEmptyHost + } + return nil +} + +func validateResolvedIP(ip net.IP) error { + // Allow loopback + if ip.IsLoopback() { + return nil + } + // Block cloud metadata and link-local + if isCloudMetadataIP(ip) || ip.IsLinkLocalUnicast() { + return ErrDeniedIP + } + return nil +} +``` + +### 3.6 Data Cleanup: Reset Stale Failure Counts + +After deploying the port fix (if not already deployed), run a one-time DB cleanup: + +```sql +-- Reset failure counts for hosts/monitors stuck from the port mismatch era +-- Only reset monitors with elevated failure counts AND no recent successful heartbeat +UPDATE uptime_hosts SET failure_count = 0, status = 'pending' WHERE status = 'down'; +UPDATE uptime_monitors SET failure_count = 0, status = 'pending' +WHERE status = 'down' + AND failure_count > 5 + AND id NOT IN ( + SELECT DISTINCT monitor_id FROM uptime_heartbeats + WHERE status = 'up' AND created_at > datetime('now', '-24 hours') + ); +``` + +This could be automated in `SyncMonitors()` or done via a migration. + +--- + +## 4. Data Flow Diagrams + +### Current Flow (Buggy) + +``` +[Proxy Host Created] → (no uptime action) + → [Wait up to 60s for ticker] + → SyncMonitors() creates monitor (status: "pending") + → CheckAll() runs: + → checkAllHosts() TCP to ForwardHost:ForwardPort + → If host up → checkMonitor() HTTP to public URL + → DB updated + → [Wait up to 30s for frontend poll] + → Frontend displays status +``` + +### Proposed Flow (Fixed) + +``` +[Proxy Host Created] + → SyncMonitors() or SyncAndCheckForHost() immediately + → Monitor created (status: "pending") + → Frontend shows "PENDING" (yellow indicator) + → Immediate checkMonitor() in background goroutine + → DB updated (status: "up" or "down") + → Frontend polls in 30s → shows actual status +``` + +--- + +## 5. Implementation Plan + +### Phase 1: Playwright E2E Tests (Behavior Specification) + +Define expected behavior before implementation: + +| Test | Description | +|------|-------------| +| New proxy host monitor appears immediately | After creating a proxy host, navigate to Uptime page, verify the monitor card exists | +| New monitor shows pending state | Verify "PENDING" badge before first check completes | +| Monitor status updates after check | Trigger manual check, verify status changes from pending/down to up | +| Verify no false DOWN on first load | Create host, wait for background check, verify status is UP (not DOWN) | + +**Files**: `tests/monitoring/uptime-monitoring.spec.ts` (extend existing suite) + +### Phase 2: Backend — Consolidate UptimeService Instance + +1. Remove second `NewUptimeService` call at `routes.go` line 414 +2. Pass `uptimeService` (line 226) to `NewUptimeHandler()` +3. Verify all handler operations use the shared instance +4. Update existing tests that may create multiple instances + +**Files**: `backend/internal/api/routes/routes.go` + +### Phase 3: Backend — Immediate Monitor Lifecycle + +1. In `ProxyHostHandler.Create()`, after saving host: call `SyncMonitors()` or create a targeted `SyncAndCheckForHost()` method +2. Add `CheckAll()` call after initial `SyncMonitors()` in the background goroutine +3. Consider adding a `SyncAndCheckForHost(hostID uint)` method to `UptimeService` that: + - Finds or creates the monitor for the given proxy host + - Immediately runs `checkMonitor()` in a goroutine + - Returns the monitor ID for the caller + +**Files**: `backend/internal/services/uptime_service.go`, `backend/internal/api/handlers/proxy_host_handler.go`, `backend/internal/api/routes/routes.go` + +### Phase 4: Frontend — Pending State Display + +1. Add `isPending` check in `MonitorCard` component +2. Add yellow/gray styling for pending state +3. Add pulsing animation for pending badge +4. Add i18n key `uptime.pending` → "CHECKING..." for **all 5 supported languages** (not just the default locale) +5. Ensure heartbeat bar handles zero-length history gracefully + +**Files**: `frontend/src/pages/Uptime.tsx`, `frontend/src/i18n/` locale files + +### Phase 5: Backend — Optional `check_url` Field (Enhancement) + +1. Add `CheckURL` field to `UptimeMonitor` model +2. Update `checkMonitor()` to use `CheckURL` if set +3. Update `SyncMonitors()` — do NOT overwrite user-configured `CheckURL` +4. Update API DTOs for create/update + +**Files**: `backend/internal/models/uptime.go`, `backend/internal/services/uptime_service.go`, `backend/internal/api/handlers/uptime_handler.go` + +### Phase 6: Frontend — Health Check URL in Edit Modal + +1. Add optional "Health Check URL" field to `EditMonitorModal` and `CreateMonitorModal` +2. Show placeholder text: "Leave empty to use monitor URL" +3. Validate URL format on frontend + +**Files**: `frontend/src/pages/Uptime.tsx` + +### Phase 7: Testing & Validation + +1. Run existing backend test suites (2178 LOC across 3 files) +2. Add tests for: + - Single `UptimeService` instance behavior + - Immediate monitor creation on proxy host create + - `CheckURL` fallback logic + - "pending" → "up" transition +3. Add edge case tests: + - **Rapid Create-Delete**: Proxy host created and immediately deleted before `SyncAndCheckForHost` goroutine completes — goroutine should handle non-existent proxy host gracefully (no panic, no orphaned monitor) + - **Concurrent Creates**: Multiple proxy hosts created simultaneously — verify `SyncMonitors()` from Create handlers doesn't conflict with background ticker's `SyncMonitors()` (no duplicate monitors, no data races) + - **Feature Flag Toggle**: If `feature.uptime.enabled` is toggled to `false` while immediate check goroutine is running — goroutine should exit cleanly without writing stale results + - **CheckURL with redirects**: `CheckURL` that 302-redirects to a private IP — redirect target must be validated against the deny list (SSRF redirect chain) +4. Run Playwright E2E suite with Docker rebuild +5. Verify coverage thresholds + +### Phase 8: Data Cleanup Migration + +1. Add one-time migration or startup hook to reset stale `failure_count` and `status` on hosts/monitors that were stuck from the port mismatch era +2. Log the cleanup action + +--- + +## 6. EARS Requirements + +1. WHEN a new proxy host is created, THE SYSTEM SHALL create a corresponding uptime monitor within 5 seconds (not waiting for the 1-minute ticker) +2. WHEN a new uptime monitor is created, THE SYSTEM SHALL immediately trigger a health check in a background goroutine +3. WHEN a monitor has status "pending" and no heartbeat history, THE SYSTEM SHALL display a distinct visual indicator (not DOWN red) +4. WHEN the server starts, THE SYSTEM SHALL run `CheckAll()` immediately after `SyncMonitors()` (not wait for first tick) +5. THE SYSTEM SHALL use a single `UptimeService` instance for both background checks and API handlers +6. WHERE a monitor has a `check_url` configured, THE SYSTEM SHALL use it for health checks instead of the monitor URL +7. WHEN a monitor's host-level TCP check succeeds but HTTP check fails, THE SYSTEM SHALL record the specific failure reason in the heartbeat message +8. IF the uptime feature flag is disabled, THEN THE SYSTEM SHALL skip all monitor sync and check operations + +--- + +## 7. Acceptance Criteria + +### Must Have + +- [ ] WHEN a new proxy host is created, a corresponding uptime monitor exists within 5 seconds +- [ ] WHEN a new uptime monitor is created, an immediate health check runs +- [ ] WHEN a monitor has status "pending", a distinct yellow/gray visual indicator is shown (not red DOWN) +- [ ] WHEN the server starts, `CheckAll()` runs immediately after `SyncMonitors()` +- [ ] Only one `UptimeService` instance exists at runtime + +### Should Have + +- [ ] WHEN a monitor has a `check_url` configured, it is used for health checks +- [ ] WHEN a monitor's host-level TCP check succeeds but HTTP check fails, the heartbeat message contains the failure reason +- [ ] Stale `failure_count` values from the port mismatch era are reset on deployment + +### Nice to Have + +- [ ] Dedicated UI indicator for "first check in progress" (animated pulse) +- [ ] Automatic detection of health endpoints (try `/health` first, fall back to `/`) + +--- + +## 8. PR Slicing Strategy + +### Decision: 3 PRs + +**Trigger reasons**: Cross-domain changes (backend + frontend + model), independent concerns (UX fix vs backend architecture vs new feature), review size management. + +### PR-1: Backend Bug Fixes (Architecture + Lifecycle) + +**Scope**: Phases 2, 3, and initial CheckAll (Section 3.4) + +**Files**: + +- `backend/internal/api/routes/routes.go` — consolidate to single UptimeService instance, add CheckAll after initial sync +- `backend/internal/services/uptime_service.go` — add `SyncAndCheckForHost()` method +- `backend/internal/api/handlers/proxy_host_handler.go` — call SyncAndCheckForHost on Create +- Backend test files — update for single instance, add new lifecycle tests +- Data cleanup migration +- `ARCHITECTURE.md` — update to reflect the UptimeService singleton consolidation (architecture change) + +**Dependencies**: None (independent of frontend changes) + +**Validation**: All backend tests pass, no duplicate UptimeService instantiation, new proxy hosts get immediate monitors, ARCHITECTURE.md reflects current design + +**Rollback**: Revert commit; behavior returns to previous (ticker-based) lifecycle + +### PR-2: Frontend Pending State + +**Scope**: Phase 4 + +**Files**: + +- `frontend/src/pages/Uptime.tsx` — add pending state handling +- `frontend/src/i18n/` locale files — add `uptime.pending` key +- `frontend/src/pages/__tests__/Uptime.spec.tsx` — update tests + +**Dependencies**: Works independently of PR-1 (pending state display improves UX regardless of backend fix timing) + +**Validation**: Playwright E2E tests pass, pending monitors show yellow indicator + +**Rollback**: Revert commit; pending monitors display as DOWN (existing behavior) + +### PR-3: Custom Health Check URL (Enhancement) + +**Scope**: Phases 5, 6 + +**Files**: + +- `backend/internal/models/uptime.go` — add CheckURL field +- `backend/internal/services/uptime_service.go` — use CheckURL in checkMonitor +- `backend/internal/api/handlers/uptime_handler.go` — update DTOs +- `frontend/src/pages/Uptime.tsx` — add form field +- Test files — add coverage for CheckURL logic + +**Dependencies**: PR-1 should be merged first (shared instance simplifies testing) + +**Validation**: Create monitor with custom health URL, verify check uses it + +**Rollback**: Revert commit; GORM auto-migration adds the column but it remains unused + +--- + +## 9. Risk Assessment + +| Risk | Severity | Likelihood | Mitigation | +|------|----------|------------|------------| +| Consolidating UptimeService instance introduces race conditions | High | Low | Existing mutex protections are designed for shared use; run race tests with `-race` flag | +| Immediate SyncMonitors on proxy host create adds latency to API response | Medium | Medium | Run SyncAndCheckForHost in a goroutine; return HTTP 201 immediately | +| "pending" UI state confuses users who expect UP/DOWN binary | Low | Low | Clear tooltip/label: "Initial health check in progress..." | +| CheckURL allows SSRF if user provides malicious URL | High | Low | Layered SSRF defense (see Section 3.5.1): write-time validation (scheme, length, parse), check-time re-validation, DNS rebinding protection (pin resolved IP against deny list), redirect chain validation. Allow loopback for self-referencing checks; block cloud metadata IPs (`169.254.169.254`, `fd00::`, etc.) | +| Data cleanup migration resets legitimate DOWN status | Medium | Medium | Only reset monitors with elevated failure counts AND no recent successful heartbeat | +| Self-referencing check (Charon) still fails due to Docker DNS | Medium | High | **PR-3 scope**: When `SyncMonitors()` creates a monitor, if `ForwardHost` resolves to loopback (`localhost`, `127.0.0.1`, or the container's own hostname), automatically set `CheckURL` to `http://{ForwardHost}:{ForwardPort}/` to bypass the DNS/Caddy round-trip. Tracked as technical debt if deferred beyond PR-3 | + +--- + +## 10. Validation Plan (Mandatory Sequence) + +0. **E2E environment prerequisite** - Determine rebuild necessity per testing policy: if application/runtime or Docker input changes are present, rebuild is required. - If rebuild is required or the container is unhealthy, run `.github/skills/scripts/skill-runner.sh docker-rebuild-e2e`. - Record container health outcome before executing tests. -1. Playwright first - - Run targeted Proxy Host dropdown and create/edit persistence scenarios. -2. Local patch coverage preflight + +1. **Playwright first** + - Run targeted uptime monitoring E2E scenarios. + +2. **Local patch coverage preflight** - Generate `test-results/local-patch-report.md` and `test-results/local-patch-report.json`. -3. Unit and coverage + +3. **Unit and coverage** - Backend coverage run (threshold >= 85%). - Frontend coverage run (threshold >= 85%). -4. Type checks + +4. **Race condition tests** + - Run `go test -race ./backend/internal/services/...` to verify single-instance thread safety. + +5. **Type checks** - Frontend TypeScript check. -5. Pre-commit + +6. **Pre-commit** - `pre-commit run --all-files` with zero blocking failures. -6. Security scans + +7. **Security scans** - CodeQL Go + JS (security-and-quality). - - Findings check gate. + - GORM security scan (model changes in PR-3). - Trivy scan. - - Conditional GORM security scan if model/DB-layer changes are made. -7. Build verification + +8. **Build verification** - Backend build + frontend build pass. -## 12. File Review: `.gitignore`, `codecov.yml`, `.dockerignore`, `Dockerfile` +--- -Assessment for this hotfix: -1. `.gitignore`: no required change for ACL/Security Headers hotfix. -2. `codecov.yml`: no required change; current exclusions/thresholds are compatible. -3. `.dockerignore`: no required change unless new hotfix-only artifact paths are introduced. -4. `Dockerfile`: no required change; incident is application logic/UI binding, not image build pipeline. +## 11. Architecture Reference -If implementation introduces new persistent test artifacts, update ignore files in the same PR. +### Two-Level Check System -## 13. Rollback and Contingency +``` +Level 1: Host-Level TCP Pre-Check +├── Purpose: Quickly determine if backend host/container is reachable +├── Method: TCP connection to ForwardHost:ForwardPort +├── Runs: Once per unique UptimeHost +├── If DOWN → Skip all Level 2 checks, mark all monitors DOWN +└── If UP → Proceed to Level 2 -1. If hotfix causes regression in proxy host save flow, revert hotfix commit and redeploy prior stable build. -2. If frontend-only fix is insufficient, activate conditional backend phase immediately. -3. If validation gates fail on security/coverage, hold merge until fixed; no partial exception for this incident. -4. Post-rollback smoke checks: - - Create host with ACL/profile. - - Edit to different ACL/profile values. - - Clear both values to `null`. - - Verify persisted values in API response and after UI reload. +Level 2: Service-Level HTTP/TCP Check +├── Purpose: Verify specific service is responding correctly +├── Method: HTTP GET to monitor URL (or CheckURL if set) +├── Runs: Per-monitor (in parallel goroutines) +└── Accepts: 2xx, 3xx, 401, 403 as "up" +``` + +### Background Ticker Flow + +``` +Server Start → Sleep 30s → SyncMonitors() + → [PROPOSED] CheckAll() + → Start 1-minute ticker + → Each tick: SyncMonitors() → CheckAll() + → checkAllHosts() [parallel, staggered] + → Group monitors by host + → For each host: + If down → markHostMonitorsDown() + If up → checkMonitor() per monitor [parallel goroutines] +``` + +### Key Configuration Values + +| Setting | Value | Source | +|---------|-------|--------| +| `batchWindow` | 30s | `NewUptimeService()` | +| `TCPTimeout` | 10s | `NewUptimeService()` | +| `MaxRetries` (host) | 2 | `NewUptimeService()` | +| `FailureThreshold` (host) | 2 | `NewUptimeService()` | +| `CheckTimeout` | 60s | `NewUptimeService()` | +| `StaggerDelay` | 100ms | `NewUptimeService()` | +| `MaxRetries` (monitor) | 3 | `UptimeMonitor.MaxRetries` default | +| Ticker interval | 1 min | `routes.go` ticker | +| Frontend poll interval | 30s | `Uptime.tsx` refetchInterval | +| History poll interval | 60s | `MonitorCard` refetchInterval | + +--- + +## 12. Rollback and Contingency + +1. **PR-1**: If consolidating UptimeService causes regressions → revert commit; background checker and API revert to two separate instances (existing behavior). +2. **PR-2**: If pending state display causes confusion → revert commit; monitors display DOWN for pending (existing behavior). +3. **PR-3**: If CheckURL introduces SSRF or regressions → revert commit; column stays in DB but is unused. +4. **Data cleanup**: If migration resets legitimate DOWN hosts → restore from SQLite backup (standard Charon backup flow). + +Post-rollback smoke checks: +- Verify background ticker creates monitors for all proxy hosts +- Verify manual health check button produces correct status +- Verify notification batching works correctly From d77d618de0d5198adb579949c788ccac58c0939e Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 02:51:18 +0000 Subject: [PATCH 09/38] feat(uptime): add pending state handling for monitors; update translations and tests --- frontend/src/locales/de/translation.json | 4 +- frontend/src/locales/en/translation.json | 4 +- frontend/src/locales/es/translation.json | 4 +- frontend/src/locales/fr/translation.json | 4 +- frontend/src/locales/zh/translation.json | 4 +- frontend/src/pages/Uptime.tsx | 21 ++++---- frontend/src/pages/__tests__/Uptime.spec.tsx | 55 ++++++++++++++++++++ 7 files changed, 82 insertions(+), 14 deletions(-) diff --git a/frontend/src/locales/de/translation.json b/frontend/src/locales/de/translation.json index e8610749..e40b3da1 100644 --- a/frontend/src/locales/de/translation.json +++ b/frontend/src/locales/de/translation.json @@ -423,7 +423,9 @@ "triggerCheck": "Sofortige Gesundheitsprüfung auslösen", "healthCheckTriggered": "Gesundheitsprüfung ausgelöst", "monitorDeleted": "Monitor gelöscht", - "deleteConfirm": "Diesen Monitor löschen? Dies kann nicht rückgängig gemacht werden." + "deleteConfirm": "Diesen Monitor löschen? Dies kann nicht rückgängig gemacht werden.", + "pending": "PRÜFUNG...", + "pendingFirstCheck": "Warten auf erste Prüfung..." }, "domains": { "title": "Domänen", diff --git a/frontend/src/locales/en/translation.json b/frontend/src/locales/en/translation.json index f90c22c3..04eca004 100644 --- a/frontend/src/locales/en/translation.json +++ b/frontend/src/locales/en/translation.json @@ -498,7 +498,9 @@ "monitorUrl": "URL", "monitorTypeHttp": "HTTP", "monitorTypeTcp": "TCP", - "urlPlaceholder": "https://example.com or tcp://host:port" + "urlPlaceholder": "https://example.com or tcp://host:port", + "pending": "CHECKING...", + "pendingFirstCheck": "Waiting for first check..." }, "domains": { "title": "Domains", diff --git a/frontend/src/locales/es/translation.json b/frontend/src/locales/es/translation.json index 07593570..a9067bbe 100644 --- a/frontend/src/locales/es/translation.json +++ b/frontend/src/locales/es/translation.json @@ -423,7 +423,9 @@ "triggerCheck": "Activar verificación de salud inmediata", "healthCheckTriggered": "Verificación de salud activada", "monitorDeleted": "Monitor eliminado", - "deleteConfirm": "¿Eliminar este monitor? Esto no se puede deshacer." + "deleteConfirm": "¿Eliminar este monitor? Esto no se puede deshacer.", + "pending": "VERIFICANDO...", + "pendingFirstCheck": "Esperando primera verificación..." }, "domains": { "title": "Dominios", diff --git a/frontend/src/locales/fr/translation.json b/frontend/src/locales/fr/translation.json index 9853dffc..525cec3f 100644 --- a/frontend/src/locales/fr/translation.json +++ b/frontend/src/locales/fr/translation.json @@ -423,7 +423,9 @@ "triggerCheck": "Déclencher une vérification de santé immédiate", "healthCheckTriggered": "Vérification de santé déclenchée", "monitorDeleted": "Moniteur supprimé", - "deleteConfirm": "Supprimer ce moniteur? Cette action est irréversible." + "deleteConfirm": "Supprimer ce moniteur? Cette action est irréversible.", + "pending": "VÉRIFICATION...", + "pendingFirstCheck": "En attente de la première vérification..." }, "domains": { "title": "Domaines", diff --git a/frontend/src/locales/zh/translation.json b/frontend/src/locales/zh/translation.json index 09e96cdd..885d64b9 100644 --- a/frontend/src/locales/zh/translation.json +++ b/frontend/src/locales/zh/translation.json @@ -423,7 +423,9 @@ "triggerCheck": "触发即时健康检查", "healthCheckTriggered": "健康检查已触发", "monitorDeleted": "监控器已删除", - "deleteConfirm": "删除此监控器？此操作无法撤销。" + "deleteConfirm": "删除此监控器？此操作无法撤销。", + "pending": "检查中...", + "pendingFirstCheck": "等待首次检查..." }, "domains": { "title": "域名", diff --git a/frontend/src/pages/Uptime.tsx b/frontend/src/pages/Uptime.tsx index 25cd4871..6861a767 100644 --- a/frontend/src/pages/Uptime.tsx +++ b/frontend/src/pages/Uptime.tsx @@ -2,7 +2,7 @@ import { useMemo, useState, type FC, type FormEvent } from 'react'; import { useTranslation } from 'react-i18next'; import { useQuery, useMutation, useQueryClient } from '@tanstack/react-query'; import { getMonitors, getMonitorHistory, updateMonitor, deleteMonitor, checkMonitor, createMonitor, syncMonitors, UptimeMonitor } from '../api/uptime'; -import { Activity, ArrowUp, ArrowDown, Settings, X, Pause, RefreshCw, Plus } from 'lucide-react'; +import { Activity, ArrowUp, ArrowDown, Settings, X, Pause, RefreshCw, Plus, Loader } from 'lucide-react'; import { toast } from 'react-hot-toast' import { formatDistanceToNow } from 'date-fns'; @@ -64,11 +64,12 @@ const MonitorCard: FC<{ monitor: UptimeMonitor; onEdit: (monitor: UptimeMonitor) ? history.reduce((a, b) => new Date(a.created_at) > new Date(b.created_at) ? a : b) : null + const isPending = monitor.status === 'pending' && (!history || history.length === 0); const isUp = latestBeat ? latestBeat.status === 'up' : monitor.status === 'up'; const isPaused = monitor.enabled === false; return ( -
+
{/* Top Row: Name (left), Badge (center-right), Settings (right) */}

{monitor.name}

@@ -76,12 +77,14 @@ const MonitorCard: FC<{ monitor: UptimeMonitor; onEdit: (monitor: UptimeMonitor)
- {isPaused ? : isUp ? : } - {isPaused ? t('uptime.paused') : monitor.status.toUpperCase()} + : isPending + ? 'bg-amber-100 text-amber-800 dark:bg-amber-900 dark:text-amber-200 animate-pulse motion-reduce:animate-none' + : isUp + ? 'bg-green-100 text-green-800 dark:bg-green-900 dark:text-green-200' + : 'bg-red-100 text-red-800 dark:bg-red-900 dark:text-red-200' + }`} data-testid="status-badge" data-status={isPaused ? 'paused' : monitor.status} role="status" aria-label={isPaused ? t('uptime.paused') : isPending ? t('uptime.pending') : isUp ? 'UP' : 'DOWN'}> + {isPaused ? : isPending ?
diff --git a/frontend/src/pages/__tests__/Uptime.spec.tsx b/frontend/src/pages/__tests__/Uptime.spec.tsx index b86ed566..924fb785 100644 --- a/frontend/src/pages/__tests__/Uptime.spec.tsx +++ b/frontend/src/pages/__tests__/Uptime.spec.tsx @@ -230,4 +230,59 @@ describe('Uptime page', () => { expect(screen.getByText('RemoteMon')).toBeInTheDocument() expect(screen.getByText('OtherMon')).toBeInTheDocument() }) + + it('shows CHECKING... state for pending monitor with no history', async () => { + const monitor = { + id: 'm13', name: 'PendingMonitor', url: 'http://example.com', type: 'http', interval: 60, enabled: true, + status: 'pending', last_check: null, latency: 0, max_retries: 3, + } + vi.mocked(uptimeApi.getMonitors).mockResolvedValue([monitor]) + vi.mocked(uptimeApi.getMonitorHistory).mockResolvedValue([]) + + renderWithProviders() + await waitFor(() => expect(screen.getByText('PendingMonitor')).toBeInTheDocument()) + const badge = screen.getByTestId('status-badge') + expect(badge).toHaveAttribute('data-status', 'pending') + expect(badge).toHaveAttribute('role', 'status') + expect(badge.textContent).toContain('CHECKING...') + expect(badge.className).toContain('bg-amber-100') + expect(badge.className).toContain('animate-pulse') + expect(screen.getByText('Waiting for first check...')).toBeInTheDocument() + }) + + it('treats pending monitor with heartbeat history as normal (not pending)', async () => { + const monitor = { + id: 'm14', name: 'PendingWithHistory', url: 'http://example.com', type: 'http', interval: 60, enabled: true, + status: 'pending', last_check: new Date().toISOString(), latency: 10, max_retries: 3, + } + const history = [ + { id: 1, monitor_id: 'm14', status: 'up', latency: 10, message: 'OK', created_at: new Date().toISOString() }, + ] + vi.mocked(uptimeApi.getMonitors).mockResolvedValue([monitor]) + vi.mocked(uptimeApi.getMonitorHistory).mockResolvedValue(history) + + renderWithProviders() + await waitFor(() => expect(screen.getByText('PendingWithHistory')).toBeInTheDocument()) + await waitFor(() => { + const badge = screen.getByTestId('status-badge') + expect(badge.textContent).not.toContain('CHECKING...') + expect(badge.className).toContain('bg-green-100') + }) + }) + + it('shows DOWN indicator for down monitor (no regression)', async () => { + const monitor = { + id: 'm15', name: 'DownMonitor', url: 'http://example.com', type: 'http', interval: 60, enabled: true, + status: 'down', last_check: new Date().toISOString(), latency: 0, max_retries: 3, + } + vi.mocked(uptimeApi.getMonitors).mockResolvedValue([monitor]) + vi.mocked(uptimeApi.getMonitorHistory).mockResolvedValue([]) + + renderWithProviders() + await waitFor(() => expect(screen.getByText('DownMonitor')).toBeInTheDocument()) + const badge = screen.getByTestId('status-badge') + expect(badge).toHaveAttribute('data-status', 'down') + expect(badge.textContent).toContain('DOWN') + expect(badge.className).toContain('bg-red-100') + }) }) From 61b73bc57b3a284e131416563f60a1c8292cb49a Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 03:49:12 +0000 Subject: [PATCH 10/38] fix(tests): increase dashboard load time threshold to 8 seconds --- tests/core/dashboard.spec.ts | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/core/dashboard.spec.ts b/tests/core/dashboard.spec.ts index 91c717b7..9e8b1efe 100644 --- a/tests/core/dashboard.spec.ts +++ b/tests/core/dashboard.spec.ts @@ -521,7 +521,7 @@ test.describe('Dashboard', () => { * Test: Dashboard loads within acceptable time */ test('should load dashboard within 5 seconds', async ({ page }) => { - const maxDashboardLoadMs = 5000; + const maxDashboardLoadMs = 8000; const startTime = Date.now(); const deadline = startTime + maxDashboardLoadMs; const remainingTime = () => Math.max(0, deadline - Date.now()); From 6483a25555b6717b2f4b927a3be1ea16e5410947 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 03:49:20 +0000 Subject: [PATCH 11/38] chore(tests): remove deprecated proxy host dropdown tests --- tests/proxy-host-dropdown-fix.spec.ts | 186 -------------------------- 1 file changed, 186 deletions(-) delete mode 100644 tests/proxy-host-dropdown-fix.spec.ts diff --git a/tests/proxy-host-dropdown-fix.spec.ts b/tests/proxy-host-dropdown-fix.spec.ts deleted file mode 100644 index 65fa857d..00000000 --- a/tests/proxy-host-dropdown-fix.spec.ts +++ /dev/null @@ -1,186 +0,0 @@ -import { test, expect } from '@playwright/test' - -type SelectionPair = { - aclLabel: string - securityHeadersLabel: string -} - -async function dismissDomainDialog(page: import('@playwright/test').Page): Promise { - const noThanksButton = page.getByRole('button', { name: /no, thanks/i }) - if (await noThanksButton.isVisible({ timeout: 1200 }).catch(() => false)) { - await noThanksButton.click() - } -} - -async function openCreateModal(page: import('@playwright/test').Page): Promise { - const addButton = page.getByRole('button', { name: /add.*proxy.*host|create/i }).first() - await expect(addButton).toBeEnabled() - await addButton.click() - await expect(page.getByRole('dialog')).toBeVisible() -} - -async function selectFirstUsableOption( - page: import('@playwright/test').Page, - trigger: import('@playwright/test').Locator, - skipPattern: RegExp -): Promise { - await trigger.click() - const listbox = page.getByRole('listbox') - await expect(listbox).toBeVisible() - - const options = listbox.getByRole('option') - const optionCount = await options.count() - expect(optionCount).toBeGreaterThan(0) - - for (let i = 0; i < optionCount; i++) { - const option = options.nth(i) - const rawLabel = (await option.textContent())?.trim() || '' - const isDisabled = (await option.getAttribute('aria-disabled')) === 'true' - - if (isDisabled || !rawLabel || skipPattern.test(rawLabel)) { - continue - } - - await option.click() - return rawLabel - } - - throw new Error('No selectable non-default option found in dropdown') -} - -async function selectOptionByName( - page: import('@playwright/test').Page, - trigger: import('@playwright/test').Locator, - optionName: RegExp -): Promise { - await trigger.click() - const listbox = page.getByRole('listbox') - await expect(listbox).toBeVisible() - - const option = listbox.getByRole('option', { name: optionName }).first() - await expect(option).toBeVisible() - const label = ((await option.textContent()) || '').trim() - await option.click() - return label -} - -async function saveProxyHost(page: import('@playwright/test').Page): Promise { - await dismissDomainDialog(page) - - const saveButton = page - .getByTestId('proxy-host-save') - .or(page.getByRole('button', { name: /^save$/i })) - .first() - await expect(saveButton).toBeEnabled() - await saveButton.click() - - const confirmSave = page.getByRole('button', { name: /yes.*save/i }).first() - if (await confirmSave.isVisible({ timeout: 1200 }).catch(() => false)) { - await confirmSave.click() - } - - await expect(page.getByRole('dialog')).not.toBeVisible({ timeout: 10000 }) -} - -async function openEditModalForDomain(page: import('@playwright/test').Page, domain: string): Promise { - const row = page.locator('tbody tr').filter({ hasText: domain }).first() - await expect(row).toBeVisible({ timeout: 10000 }) - - const editButton = row.getByRole('button', { name: /edit proxy host|edit/i }).first() - await expect(editButton).toBeVisible() - await editButton.click() - await expect(page.getByRole('dialog')).toBeVisible() -} - -async function selectNonDefaultPair( - page: import('@playwright/test').Page, - dialog: import('@playwright/test').Locator -): Promise { - const aclTrigger = dialog.getByRole('combobox', { name: /access control list/i }) - const securityHeadersTrigger = dialog.getByRole('combobox', { name: /security headers/i }) - - const aclLabel = await selectFirstUsableOption(page, aclTrigger, /no access control|public/i) - await expect(aclTrigger).toContainText(aclLabel) - - const securityHeadersLabel = await selectFirstUsableOption(page, securityHeadersTrigger, /none \(no security headers\)/i) - await expect(securityHeadersTrigger).toContainText(securityHeadersLabel) - - return { aclLabel, securityHeadersLabel } -} - -test.describe.skip('ProxyHostForm ACL/Security Headers Regression (moved to security shard)', () => { - test('should keep ACL and Security Headers behavior equivalent across create/edit flows', async ({ page }) => { - const suffix = Date.now() - const proxyName = `Dropdown Regression ${suffix}` - const proxyDomain = `dropdown-${suffix}.test.local` - - await test.step('Navigate to Proxy Hosts', async () => { - await page.goto('/proxy-hosts') - await page.waitForLoadState('networkidle') - await expect(page.getByRole('heading', { name: /proxy hosts/i })).toBeVisible() - }) - - await test.step('Create flow: select ACL + Security Headers and verify immediate form state', async () => { - await openCreateModal(page) - const dialog = page.getByRole('dialog') - - await dialog.locator('#proxy-name').fill(proxyName) - await dialog.locator('#domain-names').click() - await page.keyboard.type(proxyDomain) - await page.keyboard.press('Tab') - await dismissDomainDialog(page) - - await dialog.locator('#forward-host').fill('127.0.0.1') - await dialog.locator('#forward-port').fill('8080') - - const initialSelection = await selectNonDefaultPair(page, dialog) - - await saveProxyHost(page) - - await openEditModalForDomain(page, proxyDomain) - const reopenDialog = page.getByRole('dialog') - await expect(reopenDialog.getByRole('combobox', { name: /access control list/i })).toContainText(initialSelection.aclLabel) - await expect(reopenDialog.getByRole('combobox', { name: /security headers/i })).toContainText(initialSelection.securityHeadersLabel) - await reopenDialog.getByRole('button', { name: /cancel/i }).click() - await expect(reopenDialog).not.toBeVisible({ timeout: 5000 }) - }) - - await test.step('Edit flow: change ACL + Security Headers and verify persisted updates', async () => { - await openEditModalForDomain(page, proxyDomain) - const dialog = page.getByRole('dialog') - - const updatedSelection = await selectNonDefaultPair(page, dialog) - await saveProxyHost(page) - - await openEditModalForDomain(page, proxyDomain) - const reopenDialog = page.getByRole('dialog') - await expect(reopenDialog.getByRole('combobox', { name: /access control list/i })).toContainText(updatedSelection.aclLabel) - await expect(reopenDialog.getByRole('combobox', { name: /security headers/i })).toContainText(updatedSelection.securityHeadersLabel) - await reopenDialog.getByRole('button', { name: /cancel/i }).click() - await expect(reopenDialog).not.toBeVisible({ timeout: 5000 }) - }) - - await test.step('Edit flow: clear both to none/null and verify persisted clearing', async () => { - await openEditModalForDomain(page, proxyDomain) - const dialog = page.getByRole('dialog') - - const aclTrigger = dialog.getByRole('combobox', { name: /access control list/i }) - const securityHeadersTrigger = dialog.getByRole('combobox', { name: /security headers/i }) - - const aclNoneLabel = await selectOptionByName(page, aclTrigger, /no access control \(public\)/i) - await expect(aclTrigger).toContainText(aclNoneLabel) - - const securityNoneLabel = await selectOptionByName(page, securityHeadersTrigger, /none \(no security headers\)/i) - await expect(securityHeadersTrigger).toContainText(securityNoneLabel) - - await saveProxyHost(page) - - await openEditModalForDomain(page, proxyDomain) - const reopenDialog = page.getByRole('dialog') - await expect(reopenDialog.getByRole('combobox', { name: /access control list/i })).toContainText(/no access control \(public\)/i) - await expect(reopenDialog.getByRole('combobox', { name: /security headers/i })).toContainText(/none \(no security headers\)/i) - await reopenDialog.getByRole('button', { name: /cancel/i }).click() - await expect(reopenDialog).not.toBeVisible({ timeout: 5000 }) - }) - }) -}) From fbd94a031e59281eb36f79eaf80e06bbc7b0d855 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 03:50:43 +0000 Subject: [PATCH 12/38] fix(import): handle cancellation of stale import sessions in various states --- tests/core/caddy-import/import-page-helpers.ts | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/tests/core/caddy-import/import-page-helpers.ts b/tests/core/caddy-import/import-page-helpers.ts index 73194b45..4651579f 100644 --- a/tests/core/caddy-import/import-page-helpers.ts +++ b/tests/core/caddy-import/import-page-helpers.ts @@ -231,6 +231,10 @@ async function loginWithSetupCredentials(page: Page): Promise { } export async function resetImportSession(page: Page): Promise { + // Unconditional cancel covers sessions in any state (reviewing, pending, etc.) + await page.request.delete('/api/v1/import/cancel').catch(() => null); + await page.request.post('/api/v1/import/cancel').catch(() => null); + try { if (!page.url().includes(IMPORT_PAGE_PATH)) { await page.goto(IMPORT_PAGE_PATH, { waitUntil: 'domcontentloaded' }); @@ -329,8 +333,11 @@ export async function ensureImportFormReady(page: Page): Promise { let textareaVisible = await textarea.isVisible().catch(() => false); if (!textareaVisible) { const pendingSessionVisible = await page.getByText(/pending import session/i).first().isVisible().catch(() => false); - if (pendingSessionVisible) { - diagnosticLog('[Diag:import-ready] pending import session detected, canceling to restore textarea'); + const reviewTableVisible = await page.getByTestId('import-review-table').isVisible().catch(() => false); + if (pendingSessionVisible || reviewTableVisible) { + diagnosticLog(`[Diag:import-ready] stale session detected (pending=${pendingSessionVisible}, review=${reviewTableVisible}), canceling to restore textarea`); + await page.request.delete('/api/v1/import/cancel').catch(() => null); + await page.request.post('/api/v1/import/cancel').catch(() => null); await clearPendingImportSession(page); await page.goto(IMPORT_PAGE_PATH, { waitUntil: 'domcontentloaded' }); await assertNoAuthRedirect(page, 'ensureImportFormReady after pending-session reset'); From 09ef4f579e81cad3e0c1cb927f625a5dca8338f1 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 03:50:50 +0000 Subject: [PATCH 13/38] fix(tests): optimize response handling in Firefox import tests --- tests/core/caddy-import/caddy-import-firefox.spec.ts | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/tests/core/caddy-import/caddy-import-firefox.spec.ts b/tests/core/caddy-import/caddy-import-firefox.spec.ts index b1df798f..a8c56d25 100644 --- a/tests/core/caddy-import/caddy-import-firefox.spec.ts +++ b/tests/core/caddy-import/caddy-import-firefox.spec.ts @@ -213,10 +213,9 @@ test.describe('Caddy Import - Firefox-Specific @firefox-only', () => { await textarea.fill('cors-test.example.com { reverse_proxy localhost:3000 }'); const parseButton = page.getByRole('button', { name: /parse|review/i }); + const responsePromise = page.waitForResponse((r) => r.url().includes('/api/v1/import/upload'), { timeout: 5000 }); await parseButton.click(); - - // Wait for response - await page.waitForResponse((r) => r.url().includes('/api/v1/import/upload'), { timeout: 5000 }); + await responsePromise; // Verify no CORS issues expect(corsIssues).toHaveLength(0); From 63c9976e5f4e7c5258afaaf1a40d643f08ca8627 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 03:54:45 +0000 Subject: [PATCH 14/38] fix(tests): improve login handling in navigation tests to manage transient 401 errors --- tests/core/navigation.spec.ts | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) diff --git a/tests/core/navigation.spec.ts b/tests/core/navigation.spec.ts index 43ae1772..18f7d796 100644 --- a/tests/core/navigation.spec.ts +++ b/tests/core/navigation.spec.ts @@ -17,14 +17,22 @@ import { waitForLoadingComplete } from '../utils/wait-helpers'; test.describe('Navigation', () => { test.beforeEach(async ({ page, adminUser }) => { - await loginUser(page, adminUser); + try { + await loginUser(page, adminUser); + } catch { + // Transient 401 under full-suite load — stored auth state is still valid + } await waitForLoadingComplete(page); await page.goto('/'); await waitForLoadingComplete(page); if (page.url().includes('/login')) { - await loginUser(page, adminUser); + try { + await loginUser(page, adminUser); + } catch { + // Fall through — page retains setup auth state from storageState fixture + } await waitForLoadingComplete(page); await page.goto('/'); await waitForLoadingComplete(page); @@ -42,7 +50,11 @@ test.describe('Navigation', () => { await test.step('Verify navigation menu exists', async () => { const nav = page.getByRole('navigation'); if (!await nav.first().isVisible().catch(() => false)) { - await loginUser(page, adminUser); + try { + await loginUser(page, adminUser); + } catch { + // Stored auth state fallback + } await waitForLoadingComplete(page); await page.goto('/'); await waitForLoadingComplete(page); From 94356e7d4ecac76c31f2c08a8fcdc797d03d4e2d Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 03:56:41 +0000 Subject: [PATCH 15/38] fix(logging): convert hostID to string for improved logging in SyncAndCheckForHost --- backend/internal/services/uptime_service.go | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/backend/internal/services/uptime_service.go b/backend/internal/services/uptime_service.go index 33030392..91422a86 100644 --- a/backend/internal/services/uptime_service.go +++ b/backend/internal/services/uptime_service.go @@ -8,6 +8,7 @@ import ( "net" "net/http" "net/url" + "strconv" "strings" "sync" "time" @@ -1217,7 +1218,8 @@ func (s *UptimeService) SyncAndCheckForHost(hostID uint) { // response and this goroutine executing. var host models.ProxyHost if err := s.DB.Where("id = ?", hostID).First(&host).Error; err != nil { - logger.Log().WithField("host_id", hostID).Debug("SyncAndCheckForHost: proxy host not found (may have been deleted)") + hostIDStr := strconv.FormatUint(uint64(hostID), 10) + logger.Log().WithField("host_id", hostIDStr).Debug("SyncAndCheckForHost: proxy host not found (may have been deleted)") return } From 404aa92ea0c5dd11088fbb8d7368976d7f251408 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 05:11:18 +0000 Subject: [PATCH 16/38] fix(tests): improve response handling and session management in import tests --- .../caddy-import/caddy-import-firefox.spec.ts | 5 +++-- tests/core/caddy-import/import-page-helpers.ts | 11 ++--------- tests/core/dashboard.spec.ts | 2 +- tests/core/navigation.spec.ts | 18 +++--------------- 4 files changed, 9 insertions(+), 27 deletions(-) diff --git a/tests/core/caddy-import/caddy-import-firefox.spec.ts b/tests/core/caddy-import/caddy-import-firefox.spec.ts index a8c56d25..b1df798f 100644 --- a/tests/core/caddy-import/caddy-import-firefox.spec.ts +++ b/tests/core/caddy-import/caddy-import-firefox.spec.ts @@ -213,9 +213,10 @@ test.describe('Caddy Import - Firefox-Specific @firefox-only', () => { await textarea.fill('cors-test.example.com { reverse_proxy localhost:3000 }'); const parseButton = page.getByRole('button', { name: /parse|review/i }); - const responsePromise = page.waitForResponse((r) => r.url().includes('/api/v1/import/upload'), { timeout: 5000 }); await parseButton.click(); - await responsePromise; + + // Wait for response + await page.waitForResponse((r) => r.url().includes('/api/v1/import/upload'), { timeout: 5000 }); // Verify no CORS issues expect(corsIssues).toHaveLength(0); diff --git a/tests/core/caddy-import/import-page-helpers.ts b/tests/core/caddy-import/import-page-helpers.ts index 4651579f..73194b45 100644 --- a/tests/core/caddy-import/import-page-helpers.ts +++ b/tests/core/caddy-import/import-page-helpers.ts @@ -231,10 +231,6 @@ async function loginWithSetupCredentials(page: Page): Promise { } export async function resetImportSession(page: Page): Promise { - // Unconditional cancel covers sessions in any state (reviewing, pending, etc.) - await page.request.delete('/api/v1/import/cancel').catch(() => null); - await page.request.post('/api/v1/import/cancel').catch(() => null); - try { if (!page.url().includes(IMPORT_PAGE_PATH)) { await page.goto(IMPORT_PAGE_PATH, { waitUntil: 'domcontentloaded' }); @@ -333,11 +329,8 @@ export async function ensureImportFormReady(page: Page): Promise { let textareaVisible = await textarea.isVisible().catch(() => false); if (!textareaVisible) { const pendingSessionVisible = await page.getByText(/pending import session/i).first().isVisible().catch(() => false); - const reviewTableVisible = await page.getByTestId('import-review-table').isVisible().catch(() => false); - if (pendingSessionVisible || reviewTableVisible) { - diagnosticLog(`[Diag:import-ready] stale session detected (pending=${pendingSessionVisible}, review=${reviewTableVisible}), canceling to restore textarea`); - await page.request.delete('/api/v1/import/cancel').catch(() => null); - await page.request.post('/api/v1/import/cancel').catch(() => null); + if (pendingSessionVisible) { + diagnosticLog('[Diag:import-ready] pending import session detected, canceling to restore textarea'); await clearPendingImportSession(page); await page.goto(IMPORT_PAGE_PATH, { waitUntil: 'domcontentloaded' }); await assertNoAuthRedirect(page, 'ensureImportFormReady after pending-session reset'); diff --git a/tests/core/dashboard.spec.ts b/tests/core/dashboard.spec.ts index 9e8b1efe..91c717b7 100644 --- a/tests/core/dashboard.spec.ts +++ b/tests/core/dashboard.spec.ts @@ -521,7 +521,7 @@ test.describe('Dashboard', () => { * Test: Dashboard loads within acceptable time */ test('should load dashboard within 5 seconds', async ({ page }) => { - const maxDashboardLoadMs = 8000; + const maxDashboardLoadMs = 5000; const startTime = Date.now(); const deadline = startTime + maxDashboardLoadMs; const remainingTime = () => Math.max(0, deadline - Date.now()); diff --git a/tests/core/navigation.spec.ts b/tests/core/navigation.spec.ts index 18f7d796..43ae1772 100644 --- a/tests/core/navigation.spec.ts +++ b/tests/core/navigation.spec.ts @@ -17,22 +17,14 @@ import { waitForLoadingComplete } from '../utils/wait-helpers'; test.describe('Navigation', () => { test.beforeEach(async ({ page, adminUser }) => { - try { - await loginUser(page, adminUser); - } catch { - // Transient 401 under full-suite load — stored auth state is still valid - } + await loginUser(page, adminUser); await waitForLoadingComplete(page); await page.goto('/'); await waitForLoadingComplete(page); if (page.url().includes('/login')) { - try { - await loginUser(page, adminUser); - } catch { - // Fall through — page retains setup auth state from storageState fixture - } + await loginUser(page, adminUser); await waitForLoadingComplete(page); await page.goto('/'); await waitForLoadingComplete(page); @@ -50,11 +42,7 @@ test.describe('Navigation', () => { await test.step('Verify navigation menu exists', async () => { const nav = page.getByRole('navigation'); if (!await nav.first().isVisible().catch(() => false)) { - try { - await loginUser(page, adminUser); - } catch { - // Stored auth state fallback - } + await loginUser(page, adminUser); await waitForLoadingComplete(page); await page.goto('/'); await waitForLoadingComplete(page); From 43a63007a7a6f6e7b01dd2cb88d76b44d9a97e78 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 05:14:59 +0000 Subject: [PATCH 17/38] fix(tests): update testIgnore patterns to exclude specific caddy-import tests --- playwright.config.js | 28 +++++++++++++++++++++++++--- 1 file changed, 25 insertions(+), 3 deletions(-) diff --git a/playwright.config.js b/playwright.config.js index 1c6cd9ee..a89e353f 100644 --- a/playwright.config.js +++ b/playwright.config.js @@ -270,7 +270,15 @@ export default defineConfig({ }, dependencies: browserDependencies, testMatch: /.*\.spec\.(ts|js)$/, - testIgnore: ['**/frontend/**', '**/node_modules/**', '**/backend/**', '**/security-enforcement/**', '**/security/**'], + testIgnore: [ + '**/frontend/**', + '**/node_modules/**', + '**/backend/**', + '**/security-enforcement/**', + '**/security/**', + '**/tests/core/caddy-import/caddy-import-firefox.spec.ts', + '**/tests/core/caddy-import/caddy-import-webkit.spec.ts', + ], }, { @@ -281,7 +289,14 @@ export default defineConfig({ }, dependencies: browserDependencies, testMatch: /.*\.spec\.(ts|js)$/, - testIgnore: ['**/frontend/**', '**/node_modules/**', '**/backend/**', '**/security-enforcement/**', '**/security/**'], + testIgnore: [ + '**/frontend/**', + '**/node_modules/**', + '**/backend/**', + '**/security-enforcement/**', + '**/security/**', + '**/tests/core/caddy-import/caddy-import-webkit.spec.ts', + ], }, { @@ -292,7 +307,14 @@ export default defineConfig({ }, dependencies: browserDependencies, testMatch: /.*\.spec\.(ts|js)$/, - testIgnore: ['**/frontend/**', '**/node_modules/**', '**/backend/**', '**/security-enforcement/**', '**/security/**'], + testIgnore: [ + '**/frontend/**', + '**/node_modules/**', + '**/backend/**', + '**/security-enforcement/**', + '**/security/**', + '**/tests/core/caddy-import/caddy-import-firefox.spec.ts', + ], }, /* Test against mobile viewports. */ From ef8f23723327b361253baf01df378ef47bccdfd5 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 05:16:27 +0000 Subject: [PATCH 18/38] fix(tests): remove redundant Firefox-only test skipping logic --- tests/core/caddy-import/caddy-import-firefox.spec.ts | 8 -------- 1 file changed, 8 deletions(-) diff --git a/tests/core/caddy-import/caddy-import-firefox.spec.ts b/tests/core/caddy-import/caddy-import-firefox.spec.ts index b1df798f..6632d147 100644 --- a/tests/core/caddy-import/caddy-import-firefox.spec.ts +++ b/tests/core/caddy-import/caddy-import-firefox.spec.ts @@ -22,10 +22,6 @@ import { test, expect } from '../../fixtures/auth-fixtures'; import { Page } from '@playwright/test'; import { ensureImportUiPreconditions, resetImportSession, waitForSuccessfulImportResponse } from './import-page-helpers'; -function firefoxOnly(browserName: string) { - test.skip(browserName !== 'firefox', 'This suite only runs on Firefox'); -} - /** * Helper to set up import API mocks */ @@ -91,10 +87,6 @@ async function setupImportMocks(page: Page, success: boolean = true) { } test.describe('Caddy Import - Firefox-Specific @firefox-only', () => { - test.beforeEach(async ({ browserName }) => { - firefoxOnly(browserName); - }); - /** * TEST 1: Event listener attachment verification * Ensures the Parse button has proper click handlers in Firefox From 3cc979f5b8b7e2ad47ce686c2fd801b36a624f18 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 05:16:38 +0000 Subject: [PATCH 19/38] fix(tests): remove webkit-only test skipping logic for improved test execution --- tests/core/caddy-import/caddy-import-webkit.spec.ts | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/tests/core/caddy-import/caddy-import-webkit.spec.ts b/tests/core/caddy-import/caddy-import-webkit.spec.ts index 860dab95..2b6dfe8e 100644 --- a/tests/core/caddy-import/caddy-import-webkit.spec.ts +++ b/tests/core/caddy-import/caddy-import-webkit.spec.ts @@ -27,10 +27,6 @@ import { waitForSuccessfulImportResponse, } from './import-page-helpers'; -function webkitOnly(browserName: string) { - test.skip(browserName !== 'webkit', 'This suite only runs on WebKit'); -} - const WEBKIT_TEST_EMAIL = process.env.E2E_TEST_EMAIL || 'e2e-test@example.com'; const WEBKIT_TEST_PASSWORD = process.env.E2E_TEST_PASSWORD || 'TestPassword123!'; @@ -151,8 +147,7 @@ async function setupImportMocks(page: Page, success: boolean = true) { test.describe('Caddy Import - WebKit-Specific @webkit-only', () => { const diagnosticsByPage = new WeakMap void>(); - test.beforeEach(async ({ browserName, page, adminUser }) => { - webkitOnly(browserName); + test.beforeEach(async ({ page, adminUser }) => { diagnosticsByPage.set(page, attachImportDiagnostics(page, 'caddy-import-webkit')); await setupImportMocks(page); await ensureWebkitAuthSession(page); From 61bb19e6f32aaebd61054b6bcfbd23aeef479c70 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 05:18:33 +0000 Subject: [PATCH 20/38] fix(tests): enhance session resume handling in import tests for improved reliability --- .../caddy-import/caddy-import-gaps.spec.ts | 75 ++++++++++++++++++- 1 file changed, 71 insertions(+), 4 deletions(-) diff --git a/tests/core/caddy-import/caddy-import-gaps.spec.ts b/tests/core/caddy-import/caddy-import-gaps.spec.ts index 79fa8c52..86ecf160 100644 --- a/tests/core/caddy-import/caddy-import-gaps.spec.ts +++ b/tests/core/caddy-import/caddy-import-gaps.spec.ts @@ -473,23 +473,90 @@ test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { }); test('4.2: should restore review table with previous content when clicking Review Changes', async ({ page, testData }) => { - // SKIP: Browser-uploaded import sessions are transient (file-based only) and not persisted - // to the database. Session resume only works for Docker-mounted Caddyfiles. - // See test 4.1 skip reason for details. const domain = generateDomain(testData, 'review-changes-test'); const caddyfile = `${domain} { reverse_proxy localhost:5000 }`; + let resumeSessionId = ''; + let shouldMockPendingStatus = false; + + await page.route('**/api/v1/import/status', async (route) => { + if (!shouldMockPendingStatus || !resumeSessionId) { + await route.continue(); + return; + } + + await route.fulfill({ + status: 200, + contentType: 'application/json', + body: JSON.stringify({ + has_pending: true, + session: { + id: resumeSessionId, + state: 'reviewing', + created_at: new Date().toISOString(), + updated_at: new Date().toISOString(), + }, + }), + }); + }); + + await page.route('**/api/v1/import/preview**', async (route) => { + if (!shouldMockPendingStatus || !resumeSessionId) { + await route.continue(); + return; + } + + await route.fulfill({ + status: 200, + contentType: 'application/json', + body: JSON.stringify({ + session: { + id: resumeSessionId, + state: 'reviewing', + created_at: new Date().toISOString(), + updated_at: new Date().toISOString(), + }, + preview: { + hosts: [ + { + domain_names: domain, + forward_scheme: 'http', + forward_host: 'localhost', + forward_port: 5000, + name: domain, + }, + ], + conflicts: [], + warnings: [], + }, + caddyfile_content: caddyfile, + conflict_details: {}, + }), + }); + }); await test.step('Create import session', async () => { await page.goto('/tasks/import/caddyfile'); await fillCaddyfileTextarea(page, caddyfile); - await clickParseAndWaitForUpload(page, 'session-review-changes'); + const uploadPromise = page.waitForResponse( + r => r.url().includes('/api/v1/import/upload') && r.status() === 200, + { timeout: 15000 } + ); + await page.getByRole('button', { name: /parse|review/i }).click(); + const uploadResponse = await uploadPromise; + const uploadBody = (await uploadResponse.json().catch(() => ({}))) as { + session?: { id?: string }; + }; + resumeSessionId = uploadBody?.session?.id || ''; + expect(resumeSessionId).toBeTruthy(); await expect(page.getByTestId('import-review-table')).toBeVisible(); }); await test.step('Navigate away and back', async () => { await page.goto('/proxy-hosts'); + shouldMockPendingStatus = true; + // Wait for status API to be called after navigation const statusPromise = page.waitForResponse(r => r.url().includes('/api/v1/import/status') && r.status() === 200 From 3409e204ebed8d8bf9f4b5755c30b7b4fff33169 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 05:18:44 +0000 Subject: [PATCH 21/38] fix(tests): enhance timeout handling for UI preconditions in import page navigation --- tests/core/caddy-import/caddy-import-cross-browser.spec.ts | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/core/caddy-import/caddy-import-cross-browser.spec.ts b/tests/core/caddy-import/caddy-import-cross-browser.spec.ts index 0afa8346..703bbbd6 100644 --- a/tests/core/caddy-import/caddy-import-cross-browser.spec.ts +++ b/tests/core/caddy-import/caddy-import-cross-browser.spec.ts @@ -184,7 +184,9 @@ async function setupImportMocks( } async function gotoImportPageWithAuthRecovery(page: Page, adminUser: TestUser): Promise { - await ensureImportUiPreconditions(page, adminUser); + await expect(async () => { + await ensureImportUiPreconditions(page, adminUser); + }).toPass({ timeout: 15000 }); } test.describe('Caddy Import - Cross-Browser @cross-browser', () => { From 4ff65c83bec4faeb9560f70562ca694d592f1298 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 05:31:37 +0000 Subject: [PATCH 22/38] fix(tests): refactor CORS handling in Firefox import tests for improved clarity and reliability --- .../caddy-import/caddy-import-firefox.spec.ts | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/tests/core/caddy-import/caddy-import-firefox.spec.ts b/tests/core/caddy-import/caddy-import-firefox.spec.ts index 6632d147..c3a840e5 100644 --- a/tests/core/caddy-import/caddy-import-firefox.spec.ts +++ b/tests/core/caddy-import/caddy-import-firefox.spec.ts @@ -20,7 +20,11 @@ import { test, expect } from '../../fixtures/auth-fixtures'; import { Page } from '@playwright/test'; -import { ensureImportUiPreconditions, resetImportSession, waitForSuccessfulImportResponse } from './import-page-helpers'; +import { + ensureImportUiPreconditions, + resetImportSession, + waitForSuccessfulImportResponse, +} from './import-page-helpers'; /** * Helper to set up import API mocks @@ -205,10 +209,12 @@ test.describe('Caddy Import - Firefox-Specific @firefox-only', () => { await textarea.fill('cors-test.example.com { reverse_proxy localhost:3000 }'); const parseButton = page.getByRole('button', { name: /parse|review/i }); - await parseButton.click(); - - // Wait for response - await page.waitForResponse((r) => r.url().includes('/api/v1/import/upload'), { timeout: 5000 }); + await waitForSuccessfulImportResponse( + page, + () => parseButton.click(), + 'firefox-cors-same-origin', + /\/api\/v1\/import\/upload/i + ); // Verify no CORS issues expect(corsIssues).toHaveLength(0); From fdbba5b8388e2fc68df23d2f2f4b5ef5c9882b66 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 13:06:36 +0000 Subject: [PATCH 23/38] fix(tests): remove redundant caddy-import spec exclusions for improved test coverage --- playwright.config.js | 4 ---- 1 file changed, 4 deletions(-) diff --git a/playwright.config.js b/playwright.config.js index a89e353f..aa82818d 100644 --- a/playwright.config.js +++ b/playwright.config.js @@ -276,8 +276,6 @@ export default defineConfig({ '**/backend/**', '**/security-enforcement/**', '**/security/**', - '**/tests/core/caddy-import/caddy-import-firefox.spec.ts', - '**/tests/core/caddy-import/caddy-import-webkit.spec.ts', ], }, @@ -295,7 +293,6 @@ export default defineConfig({ '**/backend/**', '**/security-enforcement/**', '**/security/**', - '**/tests/core/caddy-import/caddy-import-webkit.spec.ts', ], }, @@ -313,7 +310,6 @@ export default defineConfig({ '**/backend/**', '**/security-enforcement/**', '**/security/**', - '**/tests/core/caddy-import/caddy-import-firefox.spec.ts', ], }, From 2204b7bd358cd78559500ac9d29e5b43e2c91616 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 13:06:47 +0000 Subject: [PATCH 24/38] fix(tests): implement retry logic for session reset and navigation stability in Caddy import tests --- .../caddy-import/caddy-import-gaps.spec.ts | 28 +++++++++++++------ 1 file changed, 20 insertions(+), 8 deletions(-) diff --git a/tests/core/caddy-import/caddy-import-gaps.spec.ts b/tests/core/caddy-import/caddy-import-gaps.spec.ts index 86ecf160..794ad4df 100644 --- a/tests/core/caddy-import/caddy-import-gaps.spec.ts +++ b/tests/core/caddy-import/caddy-import-gaps.spec.ts @@ -64,6 +64,14 @@ async function clickParseAndWaitForUpload(page: Page, context: string): Promise< } } +async function resetImportSessionWithRetry(page: Page): Promise { + // WebKit can occasionally throw a transient internal navigation error during + // route transitions; a bounded retry keeps hooks deterministic. + await expect(async () => { + await resetImportSession(page); + }).toPass({ timeout: 20000 }); +} + /** * Helper: Complete the full import flow from paste to success modal * Reusable across multiple tests to reduce duplication @@ -106,11 +114,11 @@ async function completeImportFlow( test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { test.beforeEach(async ({ page }) => { - await resetImportSession(page); + await resetImportSessionWithRetry(page); }); test.afterEach(async ({ page }) => { - await resetImportSession(page); + await resetImportSessionWithRetry(page); }); // ========================================================================= @@ -557,12 +565,16 @@ test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { await page.goto('/proxy-hosts'); shouldMockPendingStatus = true; - // Wait for status API to be called after navigation - const statusPromise = page.waitForResponse(r => - r.url().includes('/api/v1/import/status') && r.status() === 200 - ); - await page.goto('/tasks/import/caddyfile'); - await statusPromise; + // WebKit can throw a transient internal navigation error; retry deterministically. + await expect(async () => { + const statusPromise = page.waitForResponse( + r => r.url().includes('/api/v1/import/status') && r.status() === 200, + { timeout: 10000 } + ); + await page.goto('/tasks/import/caddyfile', { waitUntil: 'domcontentloaded' }); + await statusPromise; + }).toPass({ timeout: 15000 }); + await expect(page.getByTestId('import-banner')).toBeVisible({ timeout: 10000 }); }); From 739104e0294f9ecb5626a9454b272f5b30ef6b8c Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 13:14:25 +0000 Subject: [PATCH 25/38] fix(workflows): update cron schedule for weekly security rebuild and nightly promotion --- .github/workflows/security-weekly-rebuild.yml | 2 +- .github/workflows/weekly-nightly-promotion.yml | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/.github/workflows/security-weekly-rebuild.yml b/.github/workflows/security-weekly-rebuild.yml index 62e76a6c..db2916f5 100644 --- a/.github/workflows/security-weekly-rebuild.yml +++ b/.github/workflows/security-weekly-rebuild.yml @@ -6,7 +6,7 @@ name: Weekly Security Rebuild on: schedule: - - cron: '0 2 * * 0' # Sundays at 02:00 UTC + - cron: '0 12 * * 2' # Tuesdays at 12:00 UTC workflow_dispatch: inputs: force_rebuild: diff --git a/.github/workflows/weekly-nightly-promotion.yml b/.github/workflows/weekly-nightly-promotion.yml index d0f57ae4..47ad9fd6 100644 --- a/.github/workflows/weekly-nightly-promotion.yml +++ b/.github/workflows/weekly-nightly-promotion.yml @@ -5,9 +5,9 @@ name: Weekly Nightly to Main Promotion on: schedule: - # Every Monday at 10:30 UTC (5:30am EST / 6:30am EDT) + # Every Monday at 12:00 UTC (7:00am EST / 8:00am EDT) # Offset from nightly sync (09:00 UTC) to avoid schedule race and allow validation completion. - - cron: '30 10 * * 1' + - cron: '0 12 * * 1' workflow_dispatch: inputs: reason: From 67bcef32e42b5adef0225a511bf93a19c81af97c Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 13:43:42 +0000 Subject: [PATCH 26/38] fix(tests): improve header verification and response handling in Firefox import tests --- .../caddy-import/caddy-import-firefox.spec.ts | 21 ++++++++++++------- 1 file changed, 13 insertions(+), 8 deletions(-) diff --git a/tests/core/caddy-import/caddy-import-firefox.spec.ts b/tests/core/caddy-import/caddy-import-firefox.spec.ts index c3a840e5..1885da0f 100644 --- a/tests/core/caddy-import/caddy-import-firefox.spec.ts +++ b/tests/core/caddy-import/caddy-import-firefox.spec.ts @@ -245,21 +245,26 @@ test.describe('Caddy Import - Firefox-Specific @firefox-only', () => { await textarea.fill('auth-test.example.com { reverse_proxy localhost:3000 }'); const parseButton = page.getByRole('button', { name: /parse|review/i }); - await parseButton.click(); - - // Wait for request to complete - await page.waitForResponse((r) => r.url().includes('/api/v1/import/upload'), { timeout: 5000 }); + const uploadResponse = await waitForSuccessfulImportResponse( + page, + () => parseButton.click(), + 'firefox-auth-headers', + /\/api\/v1\/import\/upload/i + ); // Verify headers were captured - expect(Object.keys(requestHeaders).length).toBeGreaterThan(0); + const sentHeaders = Object.keys(requestHeaders).length > 0 + ? requestHeaders + : uploadResponse.request().headers(); + expect(Object.keys(sentHeaders).length).toBeGreaterThan(0); // Verify cookie or authorization header present - const hasCookie = !!requestHeaders['cookie']; - const hasAuth = !!requestHeaders['authorization']; + const hasCookie = !!sentHeaders['cookie']; + const hasAuth = !!sentHeaders['authorization']; expect(hasCookie || hasAuth).toBeTruthy(); // Verify content-type is correct - expect(requestHeaders['content-type']).toContain('application/json'); + expect(sentHeaders['content-type']).toContain('application/json'); }); }); From 8e1b9d91e2219263e167255e775e6e60ea940eed Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 13:43:50 +0000 Subject: [PATCH 27/38] fix(tests): enhance session handling and cleanup in Caddy import tests --- .../caddy-import/caddy-import-gaps.spec.ts | 31 ++++++++++++++++--- 1 file changed, 26 insertions(+), 5 deletions(-) diff --git a/tests/core/caddy-import/caddy-import-gaps.spec.ts b/tests/core/caddy-import/caddy-import-gaps.spec.ts index 794ad4df..e66d4870 100644 --- a/tests/core/caddy-import/caddy-import-gaps.spec.ts +++ b/tests/core/caddy-import/caddy-import-gaps.spec.ts @@ -118,7 +118,9 @@ test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { }); test.afterEach(async ({ page }) => { - await resetImportSessionWithRetry(page); + await resetImportSessionWithRetry(page).catch(() => { + // Best-effort cleanup only; preserve primary test failure signal. + }); }); // ========================================================================= @@ -399,7 +401,7 @@ test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { // Gap 4: Session Resume via Banner // ========================================================================= test.describe('Session Resume via Banner', () => { - test('4.1: should show pending session banner when returning to import page', async ({ page, testData }) => { + test('4.1: should show pending session banner when returning to import page', async ({ page, testData, browserName, adminUser }) => { const domain = generateDomain(testData, 'session-resume-test'); const caddyfile = `${domain} { reverse_proxy localhost:4000 }`; let resumeSessionId = ''; @@ -427,7 +429,12 @@ test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { }); await test.step('Create import session by parsing content', async () => { - await page.goto('/tasks/import/caddyfile'); + await page.goto('/tasks/import/caddyfile', { waitUntil: 'domcontentloaded' }); + if (browserName === 'webkit') { + await ensureAuthenticatedImportFormReady(page, adminUser); + } else { + await ensureImportFormReady(page); + } await fillCaddyfileTextarea(page, caddyfile); const uploadPromise = page.waitForResponse( @@ -478,9 +485,13 @@ test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { // Review table should NOT be visible initially (until clicking Review Changes) await expect(page.getByTestId('import-review-table')).not.toBeVisible(); }); + + await test.step('Cleanup mocked routes', async () => { + await page.unroute('**/api/v1/import/status'); + }); }); - test('4.2: should restore review table with previous content when clicking Review Changes', async ({ page, testData }) => { + test('4.2: should restore review table with previous content when clicking Review Changes', async ({ page, testData, browserName, adminUser }) => { const domain = generateDomain(testData, 'review-changes-test'); const caddyfile = `${domain} { reverse_proxy localhost:5000 }`; let resumeSessionId = ''; @@ -543,7 +554,12 @@ test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { }); await test.step('Create import session', async () => { - await page.goto('/tasks/import/caddyfile'); + await page.goto('/tasks/import/caddyfile', { waitUntil: 'domcontentloaded' }); + if (browserName === 'webkit') { + await ensureAuthenticatedImportFormReady(page, adminUser); + } else { + await ensureImportFormReady(page); + } await fillCaddyfileTextarea(page, caddyfile); const uploadPromise = page.waitForResponse( @@ -594,6 +610,11 @@ test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { // Note: Some implementations keep banner visible but change its content // If banner remains, it should show different text }); + + await test.step('Cleanup mocked routes', async () => { + await page.unroute('**/api/v1/import/status'); + await page.unroute('**/api/v1/import/preview**'); + }); }); }); From dbff270d22ec573f3a310d1e44670041cb61036d Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 14:04:40 +0000 Subject: [PATCH 28/38] fix(tests): update input handling in ProxyHostForm tests for improved reliability --- frontend/src/components/__tests__/ProxyHostForm.test.tsx | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/frontend/src/components/__tests__/ProxyHostForm.test.tsx b/frontend/src/components/__tests__/ProxyHostForm.test.tsx index 9e7f57b8..5465c8e6 100644 --- a/frontend/src/components/__tests__/ProxyHostForm.test.tsx +++ b/frontend/src/components/__tests__/ProxyHostForm.test.tsx @@ -1440,12 +1440,17 @@ describe('ProxyHostForm', () => { ) - await userEvent.type(screen.getByLabelText(/^Name/), 'Remote Mapping') - await userEvent.type(screen.getByPlaceholderText('example.com, www.example.com'), 'remote.existing.com') + fireEvent.change(screen.getByLabelText(/^Name/), { target: { value: 'Remote Mapping' } }) + fireEvent.change(screen.getByPlaceholderText('example.com, www.example.com'), { target: { value: 'remote.existing.com' } }) await selectComboboxOption('Source', 'Local Docker Registry (localhost)') await selectComboboxOption('Containers', 'remote-app (nginx:latest)') + await waitFor(() => { + expect(screen.getByLabelText(/^Host$/)).toHaveValue('localhost') + expect(screen.getByLabelText(/^Port$/)).toHaveValue(18080) + }) + await userEvent.click(screen.getByText('Save')) await waitFor(() => { From 871adca27005f04b241c1f141654fea6edcbe1b7 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 14:08:13 +0000 Subject: [PATCH 29/38] fix(deps): update modernc.org/libc to v1.69.0 for improved compatibility --- backend/go.mod | 2 +- backend/go.sum | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/backend/go.mod b/backend/go.mod index 75ec8a47..5e60f1f7 100644 --- a/backend/go.mod +++ b/backend/go.mod @@ -95,7 +95,7 @@ require ( google.golang.org/protobuf v1.36.11 // indirect gopkg.in/yaml.v3 v3.0.1 // indirect gotest.tools/v3 v3.5.2 // indirect - modernc.org/libc v1.68.1 // indirect + modernc.org/libc v1.69.0 // indirect modernc.org/mathutil v1.7.1 // indirect modernc.org/memory v1.11.0 // indirect modernc.org/sqlite v1.46.1 // indirect diff --git a/backend/go.sum b/backend/go.sum index 1fed2afc..489d36a5 100644 --- a/backend/go.sum +++ b/backend/go.sum @@ -253,8 +253,8 @@ modernc.org/gc/v3 v3.1.2 h1:ZtDCnhonXSZexk/AYsegNRV1lJGgaNZJuKjJSWKyEqo= modernc.org/gc/v3 v3.1.2/go.mod h1:HFK/6AGESC7Ex+EZJhJ2Gni6cTaYpSMmU/cT9RmlfYY= modernc.org/goabi0 v0.2.0 h1:HvEowk7LxcPd0eq6mVOAEMai46V+i7Jrj13t4AzuNks= modernc.org/goabi0 v0.2.0/go.mod h1:CEFRnnJhKvWT1c1JTI3Avm+tgOWbkOu5oPA8eH8LnMI= -modernc.org/libc v1.68.1 h1:qNL/EzzdzNicXwJ9Gj2IHlVjuqRQsPXngFRaDMGuFwE= -modernc.org/libc v1.68.1/go.mod h1:YfLLduUEbodNV2xLU5JOnRHBTAHVHsVW3bVYGw0ZCV4= +modernc.org/libc v1.69.0 h1:YQJ5QMSReTgQ3QFmI0dudfjXIjCcYTUxcH8/9P9f0D8= +modernc.org/libc v1.69.0/go.mod h1:YfLLduUEbodNV2xLU5JOnRHBTAHVHsVW3bVYGw0ZCV4= modernc.org/mathutil v1.7.1 h1:GCZVGXdaN8gTqB1Mf/usp1Y/hSqgI2vAGGP4jZMCxOU= modernc.org/mathutil v1.7.1/go.mod h1:4p5IwJITfppl0G4sUEDtCr4DthTaT47/N3aT6MhfgJg= modernc.org/memory v1.11.0 h1:o4QC8aMQzmcwCK3t3Ux/ZHmwFPzE6hf2Y5LbkRs+hbI= From 7a12ab7928040854bb1bd931dc12d521d06a1d2a Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 16:26:24 +0000 Subject: [PATCH 30/38] fix(uptime): remove redundant host failure count reset logic --- backend/internal/services/uptime_service.go | 9 --------- backend/internal/services/uptime_service_pr1_test.go | 8 ++++---- 2 files changed, 4 insertions(+), 13 deletions(-) diff --git a/backend/internal/services/uptime_service.go b/backend/internal/services/uptime_service.go index 91422a86..8ecc6d4b 100644 --- a/backend/internal/services/uptime_service.go +++ b/backend/internal/services/uptime_service.go @@ -1292,14 +1292,5 @@ func (s *UptimeService) CleanupStaleFailureCounts() error { logger.Log().WithField("reset_count", result.RowsAffected).Info("Reset stale monitor failure counts") } - hostResult := s.DB.Exec(`UPDATE uptime_hosts SET failure_count = 0, status = 'pending' WHERE status = 'down'`) - if hostResult.Error != nil { - return fmt.Errorf("cleanup stale host failure counts: %w", hostResult.Error) - } - - if hostResult.RowsAffected > 0 { - logger.Log().WithField("reset_count", hostResult.RowsAffected).Info("Reset stale host failure counts") - } - return nil } diff --git a/backend/internal/services/uptime_service_pr1_test.go b/backend/internal/services/uptime_service_pr1_test.go index 7c6b425e..6de1104c 100644 --- a/backend/internal/services/uptime_service_pr1_test.go +++ b/backend/internal/services/uptime_service_pr1_test.go @@ -311,11 +311,11 @@ func TestCleanupStaleFailureCounts_SkipsLowFailureCount(t *testing.T) { assert.Equal(t, "down", m.Status) } -func TestCleanupStaleFailureCounts_ResetsStaleHosts(t *testing.T) { +func TestCleanupStaleFailureCounts_DoesNotResetDownHosts(t *testing.T) { db := setupPR1TestDB(t) svc := NewUptimeService(db, nil) - // Create a "stuck" host + // Create a host that is currently down. host := models.UptimeHost{ ID: uuid.New().String(), Host: "stuck-host.local", @@ -330,8 +330,8 @@ func TestCleanupStaleFailureCounts_ResetsStaleHosts(t *testing.T) { var h models.UptimeHost require.NoError(t, db.First(&h, "id = ?", host.ID).Error) - assert.Equal(t, 0, h.FailureCount) - assert.Equal(t, "pending", h.Status) + assert.Equal(t, 10, h.FailureCount, "cleanup must not reset host failure_count") + assert.Equal(t, "down", h.Status, "cleanup must not reset host status") } // setupPR1ConcurrentDB creates a file-based SQLite database with WAL mode and From 6f5c8873f9b68c3edfa2cafd8501fc22b2088b2e Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 16:30:21 +0000 Subject: [PATCH 31/38] fix(tests): refactor proxy host creation to use dynamic server URLs in uptime tests --- .../services/uptime_service_pr1_test.go | 50 +++++++++++++++---- 1 file changed, 40 insertions(+), 10 deletions(-) diff --git a/backend/internal/services/uptime_service_pr1_test.go b/backend/internal/services/uptime_service_pr1_test.go index 6de1104c..dd3c97fd 100644 --- a/backend/internal/services/uptime_service_pr1_test.go +++ b/backend/internal/services/uptime_service_pr1_test.go @@ -2,8 +2,11 @@ package services import ( "fmt" + "net/http" + "net/http/httptest" "os" "path/filepath" + "strings" "sync" "testing" "time" @@ -71,6 +74,19 @@ func createTestProxyHost(t *testing.T, db *gorm.DB, name, domain, forwardHost st return host } +func createAlwaysOKServer(t *testing.T) *httptest.Server { + t.Helper() + server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + w.WriteHeader(http.StatusOK) + })) + t.Cleanup(server.Close) + return server +} + +func hostPortFromServerURL(serverURL string) string { + return strings.TrimPrefix(serverURL, "http://") +} + // --- Fix 1: Singleton UptimeService --- func TestSingletonUptimeService_SharedState(t *testing.T) { @@ -95,8 +111,10 @@ func TestSyncAndCheckForHost_CreatesMonitorAndHeartbeat(t *testing.T) { db := setupPR1TestDB(t) enableUptimeFeature(t, db) svc := NewUptimeService(db, nil) + server := createAlwaysOKServer(t) + domain := hostPortFromServerURL(server.URL) - host := createTestProxyHost(t, db, "test-host", "example.com", "192.168.1.100") + host := createTestProxyHost(t, db, "test-host", domain, "192.168.1.100") // Execute synchronously (normally called as goroutine) svc.SyncAndCheckForHost(host.ID) @@ -105,7 +123,7 @@ func TestSyncAndCheckForHost_CreatesMonitorAndHeartbeat(t *testing.T) { var monitor models.UptimeMonitor err := db.Where("proxy_host_id = ?", host.ID).First(&monitor).Error require.NoError(t, err, "monitor should be created for the proxy host") - assert.Equal(t, "http://example.com", monitor.URL) + assert.Equal(t, "http://"+domain, monitor.URL) assert.Equal(t, "192.168.1.100", monitor.UpstreamHost) assert.Contains(t, []string{"up", "down", "pending"}, monitor.Status, "status should be set by checkMonitor") @@ -119,11 +137,13 @@ func TestSyncAndCheckForHost_SSLForcedUsesHTTPS(t *testing.T) { db := setupPR1TestDB(t) enableUptimeFeature(t, db) svc := NewUptimeService(db, nil) + server := createAlwaysOKServer(t) + domain := hostPortFromServerURL(server.URL) host := models.ProxyHost{ UUID: uuid.New().String(), Name: "ssl-host", - DomainNames: "secure.example.com", + DomainNames: domain, ForwardScheme: "https", ForwardHost: "192.168.1.200", ForwardPort: 443, @@ -136,7 +156,7 @@ func TestSyncAndCheckForHost_SSLForcedUsesHTTPS(t *testing.T) { var monitor models.UptimeMonitor require.NoError(t, db.Where("proxy_host_id = ?", host.ID).First(&monitor).Error) - assert.Equal(t, "https://secure.example.com", monitor.URL) + assert.Equal(t, "https://"+domain, monitor.URL) } func TestSyncAndCheckForHost_DeletedHostNoPanic(t *testing.T) { @@ -159,8 +179,10 @@ func TestSyncAndCheckForHost_ExistingMonitorSkipsCreate(t *testing.T) { db := setupPR1TestDB(t) enableUptimeFeature(t, db) svc := NewUptimeService(db, nil) + server := createAlwaysOKServer(t) + domain := hostPortFromServerURL(server.URL) - host := createTestProxyHost(t, db, "existing-mon", "existing.com", "10.0.0.1") + host := createTestProxyHost(t, db, "existing-mon", domain, "10.0.0.1") // Pre-create a monitor existingMonitor := models.UptimeMonitor{ @@ -168,7 +190,7 @@ func TestSyncAndCheckForHost_ExistingMonitorSkipsCreate(t *testing.T) { ProxyHostID: &host.ID, Name: "pre-existing", Type: "http", - URL: "http://existing.com", + URL: "http://" + domain, Interval: 60, Enabled: true, Status: "up", @@ -195,8 +217,10 @@ func TestSyncAndCheckForHost_DisabledFeatureNoop(t *testing.T) { Category: "feature", }).Error) svc := NewUptimeService(db, nil) + server := createAlwaysOKServer(t) + domain := hostPortFromServerURL(server.URL) - host := createTestProxyHost(t, db, "disabled-host", "disabled.com", "10.0.0.2") + host := createTestProxyHost(t, db, "disabled-host", domain, "10.0.0.2") svc.SyncAndCheckForHost(host.ID) @@ -210,8 +234,10 @@ func TestSyncAndCheckForHost_MissingSetting_StillCreates(t *testing.T) { db := setupPR1TestDB(t) // No setting at all — the method should proceed (default: enabled behavior) svc := NewUptimeService(db, nil) + server := createAlwaysOKServer(t) + domain := hostPortFromServerURL(server.URL) - host := createTestProxyHost(t, db, "no-setting", "nosetting.com", "10.0.0.3") + host := createTestProxyHost(t, db, "no-setting", domain, "10.0.0.3") svc.SyncAndCheckForHost(host.ID) @@ -368,13 +394,15 @@ func TestSyncAndCheckForHost_ConcurrentCreates_NoDuplicates(t *testing.T) { db := setupPR1ConcurrentDB(t) enableUptimeFeature(t, db) svc := NewUptimeService(db, nil) + server := createAlwaysOKServer(t) + domain := hostPortFromServerURL(server.URL) // Create multiple proxy hosts with unique domains hosts := make([]models.ProxyHost, 5) for i := range hosts { hosts[i] = createTestProxyHost(t, db, fmt.Sprintf("concurrent-host-%d", i), - fmt.Sprintf("concurrent-%d.com", i), + domain, fmt.Sprintf("10.0.0.%d", 100+i), ) } @@ -401,8 +429,10 @@ func TestSyncAndCheckForHost_ConcurrentSameHost_NoDuplicates(t *testing.T) { db := setupPR1ConcurrentDB(t) enableUptimeFeature(t, db) svc := NewUptimeService(db, nil) + server := createAlwaysOKServer(t) + domain := hostPortFromServerURL(server.URL) - host := createTestProxyHost(t, db, "race-host", "race.com", "10.0.0.200") + host := createTestProxyHost(t, db, "race-host", domain, "10.0.0.200") var wg sync.WaitGroup for i := 0; i < 10; i++ { From f20e789a16559f1c315afeae803d5ae666678c74 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 16:30:51 +0000 Subject: [PATCH 32/38] fix(tests): increase timeout for ProxyHostForm tests to improve reliability --- docs/reports/qa_report.md | 125 +++++++++--------- .../__tests__/ProxyHostForm.test.tsx | 2 +- 2 files changed, 61 insertions(+), 66 deletions(-) diff --git a/docs/reports/qa_report.md b/docs/reports/qa_report.md index 77915271..b2dc9a57 100644 --- a/docs/reports/qa_report.md +++ b/docs/reports/qa_report.md @@ -1,85 +1,80 @@ -double check our caddy version# QA Report: Nightly Workflow Fix Audit +## QA Report - PR #779 -- Date: 2026-02-27 -- Scope: - - `.github/workflows/nightly-build.yml` - 1. `pr_number` failure avoidance in nightly dispatch path - 2. Deterministic Syft SBOM generation with fallback - - `.github/workflows/security-pr.yml` contract check (`pr_number` required) +- Date: 2026-03-01 +- Scope: Post-remediation merge-readiness gates after Caddy Import E2E fix -## Findings (Ordered by Severity) +## E2E Status -### ✅ No blocking findings in audited scope +- Command status provided by current PR context: + `npx playwright test --project=chromium --project=firefox --project=webkit tests/core/caddy-import` +- Result: `106 passed, 0 failed, 0 skipped` +- Gate: PASS -1. `actionlint` validation passed for modified workflow. - - Command: `actionlint .github/workflows/nightly-build.yml` - - Result: PASS (no diagnostics) +## Patch Report Status -2. `pr_number` nightly dispatch failure path is avoided by excluding PR-only workflow from nightly fan-out. - - `security-pr.yml` removed from dispatch list in `.github/workflows/nightly-build.yml:103` - - Explicit log note added at `.github/workflows/nightly-build.yml:110` +- Command: `bash scripts/local-patch-report.sh` +- Artifacts: + - `test-results/local-patch-report.md` (present) + - `test-results/local-patch-report.json` (present) +- Result: PASS (artifacts generated) +- Notes: + - Warning: overall patch coverage `81.7%` below advisory threshold `90.0%` + - Warning: backend patch coverage `81.6%` below advisory threshold `85.0%` -3. SBOM generation is now deterministic with explicit primary pin and verified fallback. - - Primary action pins Syft version at `.github/workflows/nightly-build.yml:231` - - Fallback installs pinned `v1.42.1` with checksum verification at `.github/workflows/nightly-build.yml:245` - - Mandatory artifact verification added at `.github/workflows/nightly-build.yml:268` +## Backend Coverage -4. No permission broadening in modified sections. - - Dispatch job permissions remain `actions: write`, `contents: read` at `.github/workflows/nightly-build.yml:84` - - Build job permissions remain `contents: read`, `packages: write`, `id-token: write` at `.github/workflows/nightly-build.yml:145` - - Diff review confirms no `permissions` changes in the modified hunk. +- Command: `.github/skills/scripts/skill-runner.sh test-backend-coverage` +- Result: PASS +- Metrics: + - Statement coverage: `87.5%` + - Line coverage: `87.7%` + - Gate threshold observed in run: `87%` -5. Action pinning remains SHA-based in modified sections. - - `actions/github-script` pinned SHA at `.github/workflows/nightly-build.yml:89` - - `anchore/sbom-action` pinned SHA at `.github/workflows/nightly-build.yml:226` - - `actions/upload-artifact` pinned SHA at `.github/workflows/nightly-build.yml:283` +## Frontend Coverage -6. `security-pr.yml` contract still requires `pr_number`. - - `workflow_dispatch.inputs.pr_number.required: true` at `.github/workflows/security-pr.yml:14` +- Command: `.github/skills/scripts/skill-runner.sh test-frontend-coverage` +- Result: FAIL +- Failure root cause: + - Test timeout at `frontend/src/components/__tests__/ProxyHostForm.test.tsx:1419` + - Failing test: `maps remote docker container to remote host and public port` + - Error: `Test timed out in 5000ms` +- Coverage snapshot produced before failure: + - Statements: `88.95%` + - Lines: `89.62%` + - Functions: `86.05%` + - Branches: `81.3%` -## Pass/Fail Decision +## Typecheck -- QA Status: **PASS with caveats** -- Reason: All requested static validations pass and the scoped workflow logic changes satisfy the audit requirements. +- Command: `npm --prefix frontend run type-check` +- Result: PASS -## Residual Risks +## Pre-commit -1. Fallback integrity uses checksum file from the same release origin as the tarball. - - Impact: If release origin is compromised, checksum verification alone may not detect tampering. - - Suggested hardening: verify signed release metadata or verify Syft artifact signature (Cosign/GitHub attestations) in fallback path. +- Command: `pre-commit run --all-files` +- Result: PASS +- Notable hooks: `golangci-lint (Fast Linters - BLOCKING)`, `Frontend TypeScript Check`, `Frontend Lint (Fix)` all passed -2. Runtime behavior is not fully proven by local static checks. - - Impact: Dispatch and SBOM behavior still require a real GitHub Actions run to prove end-to-end execution. +## Security Scans -## Remote Execution Limitation and Manual Verification +- Trivy filesystem scan: + - Command: `.github/skills/scripts/skill-runner.sh security-scan-trivy` + - Result: PASS + - Critical/High findings: `0/0` -I did not execute remote nightly runs for this exact local diff in this audit. Local `actionlint` and source inspection were performed. To validate end-to-end behavior on GitHub Actions, run: +- Docker image scan: + - Command: `.github/skills/scripts/skill-runner.sh security-scan-docker-image` + - Result: PASS + - Critical/High findings: `0/0` + - Additional findings: `10 medium`, `3 low` (non-blocking) -```bash -cd /projects/Charon +## Remediation Required Before Merge -# 1) Syntax/lint (already run locally) -actionlint .github/workflows/nightly-build.yml +1. Stabilize the timed-out frontend test at `frontend/src/components/__tests__/ProxyHostForm.test.tsx:1419`. +2. Re-run `.github/skills/scripts/skill-runner.sh test-frontend-coverage` until the suite is fully green. +3. Optional quality improvement: raise patch coverage warnings (`81.7%` overall, `81.6%` backend) with targeted tests on uncovered changed lines from `test-results/local-patch-report.md`. -# 2) Trigger nightly workflow (manual) -gh workflow run nightly-build.yml --ref nightly -f reason="qa-nightly-audit" -f skip_tests=true +## Final Merge Recommendation -# 3) Inspect latest nightly run -gh run list --workflow "Nightly Build & Package" --branch nightly --limit 1 -gh run view --log - -# 4) Confirm no security-pr dispatch error in nightly logs -# Expectation: no "Missing required input 'pr_number' not provided" - -# 5) Confirm security-pr contract still enforced -gh workflow run security-pr.yml --ref nightly -# Expectation: dispatch rejected due to required missing input pr_number - -# 6) Positive contract check with explicit pr_number -gh workflow run security-pr.yml --ref nightly -f pr_number= -``` - -Expected outcomes: -- Nightly run completes dispatch phase without `pr_number` input failure. -- SBOM generation succeeds via primary or fallback path and uploads `sbom-nightly.json`. -- `security-pr.yml` continues enforcing required `pr_number` for manual dispatch. +- Recommendation: **NO-GO** +- Reason: Required frontend coverage gate did not pass due to a deterministic test timeout. diff --git a/frontend/src/components/__tests__/ProxyHostForm.test.tsx b/frontend/src/components/__tests__/ProxyHostForm.test.tsx index 5465c8e6..c579f072 100644 --- a/frontend/src/components/__tests__/ProxyHostForm.test.tsx +++ b/frontend/src/components/__tests__/ProxyHostForm.test.tsx @@ -1459,7 +1459,7 @@ describe('ProxyHostForm', () => { forward_port: 18080, })) }) - }) + }, 15000) it('updates domain using selected container when base domain changes', async () => { const { useDocker } = await import('../../hooks/useDocker') From 0241de69f476a528adf3bede960342843c03dda2 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 16:33:09 +0000 Subject: [PATCH 33/38] fix(uptime): enhance monitor status handling and display logic in MonitorCard --- frontend/src/pages/Uptime.tsx | 38 ++++++++++++++++++++++++++--------- 1 file changed, 28 insertions(+), 10 deletions(-) diff --git a/frontend/src/pages/Uptime.tsx b/frontend/src/pages/Uptime.tsx index 6861a767..8bbcfada 100644 --- a/frontend/src/pages/Uptime.tsx +++ b/frontend/src/pages/Uptime.tsx @@ -6,6 +6,18 @@ import { Activity, ArrowUp, ArrowDown, Settings, X, Pause, RefreshCw, Plus, Load import { toast } from 'react-hot-toast' import { formatDistanceToNow } from 'date-fns'; +type BaseMonitorStatus = 'up' | 'down' | 'pending'; +type EffectiveMonitorStatus = BaseMonitorStatus | 'paused'; + +const normalizeMonitorStatus = (status: string | undefined): BaseMonitorStatus => { + const normalized = status?.toLowerCase(); + if (normalized === 'up' || normalized === 'down' || normalized === 'pending') { + return normalized; + } + + return 'down'; +}; + const MonitorCard: FC<{ monitor: UptimeMonitor; onEdit: (monitor: UptimeMonitor) => void; t: (key: string, options?: Record) => string }> = ({ monitor, onEdit, t }) => { const { data: history } = useQuery({ queryKey: ['uptimeHistory', monitor.id], @@ -64,27 +76,33 @@ const MonitorCard: FC<{ monitor: UptimeMonitor; onEdit: (monitor: UptimeMonitor) ? history.reduce((a, b) => new Date(a.created_at) > new Date(b.created_at) ? a : b) : null - const isPending = monitor.status === 'pending' && (!history || history.length === 0); - const isUp = latestBeat ? latestBeat.status === 'up' : monitor.status === 'up'; + const hasHistory = Boolean(history && history.length > 0); const isPaused = monitor.enabled === false; + const effectiveStatus: EffectiveMonitorStatus = isPaused + ? 'paused' + : latestBeat + ? (latestBeat.status === 'up' ? 'up' : 'down') + : monitor.status === 'pending' && !hasHistory + ? 'pending' + : normalizeMonitorStatus(monitor.status); return ( -
+
{/* Top Row: Name (left), Badge (center-right), Settings (right) */}

{monitor.name}

- {isPaused ? : isPending ?
From d94c9ba623fdc02c4dcbf8acc1fc6975c91c331e Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 17:17:49 +0000 Subject: [PATCH 34/38] fix(tests): enhance overwrite resolution flow test to handle browser-specific authentication --- tests/core/caddy-import/caddy-import-gaps.spec.ts | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/tests/core/caddy-import/caddy-import-gaps.spec.ts b/tests/core/caddy-import/caddy-import-gaps.spec.ts index e66d4870..ac64fe9c 100644 --- a/tests/core/caddy-import/caddy-import-gaps.spec.ts +++ b/tests/core/caddy-import/caddy-import-gaps.spec.ts @@ -328,7 +328,7 @@ test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { // Gap 3: Overwrite Resolution Flow // ========================================================================= test.describe('Overwrite Resolution Flow', () => { - test('3.1: should update existing host when selecting Replace with Imported resolution', async ({ page, request, testData }) => { + test('3.1: should update existing host when selecting Replace with Imported resolution', async ({ page, request, testData, browserName, adminUser }) => { // Create existing host with initial config const result = await testData.createProxyHost({ domain: 'overwrite-test.example.com', @@ -341,6 +341,11 @@ test.describe('Caddy Import Gap Coverage @caddy-import-gaps', () => { await test.step('Navigate to import page and parse conflicting Caddyfile', async () => { await page.goto('/tasks/import/caddyfile'); + if (browserName === 'webkit') { + await ensureAuthenticatedImportFormReady(page, adminUser); + } else { + await ensureImportFormReady(page); + } // Import with different config (new-server:9000) const caddyfile = `${namespacedDomain} { reverse_proxy new-server:9000 }`; await fillCaddyfileTextarea(page, caddyfile); From f79f0218c5a5b97cf3905dbbab2e8ad9a50d283b Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Sun, 1 Mar 2026 17:38:01 +0000 Subject: [PATCH 35/38] fix(tests): update mock heartbeat generation to align with monitor's latest status --- tests/monitoring/uptime-monitoring.spec.ts | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/tests/monitoring/uptime-monitoring.spec.ts b/tests/monitoring/uptime-monitoring.spec.ts index 10c8c0cd..34a26361 100644 --- a/tests/monitoring/uptime-monitoring.spec.ts +++ b/tests/monitoring/uptime-monitoring.spec.ts @@ -93,11 +93,16 @@ const mockMonitors: UptimeMonitor[] = [ /** * Generate mock heartbeat history */ -const generateMockHistory = (monitorId: string, count: number = 60): UptimeHeartbeat[] => { +const generateMockHistory = ( + monitorId: string, + count: number = 60, + latestStatus: 'up' | 'down' = 'up' +): UptimeHeartbeat[] => { return Array.from({ length: count }, (_, i) => ({ id: i, monitor_id: monitorId, - status: i % 5 === 0 ? 'down' : 'up', + // Keep the newest heartbeat aligned with the monitor's expected current state. + status: i === 0 ? latestStatus : i % 5 === 0 ? 'down' : 'up', latency: Math.floor(Math.random() * 100), message: 'OK', created_at: new Date(Date.now() - i * 60000).toISOString(), @@ -180,7 +185,8 @@ async function setupMonitorsWithHistory( await setupMonitorsAPI(page, monitors); for (const monitor of monitors) { - const history = generateMockHistory(monitor.id, 60); + const latestStatus = monitor.status === 'down' ? 'down' : 'up'; + const history = generateMockHistory(monitor.id, 60, latestStatus); await setupHistoryAPI(page, monitor.id, history); } } From aaddb884883d46e1e3e736690e4e222b858458c9 Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Mon, 2 Mar 2026 00:24:03 +0000 Subject: [PATCH 36/38] fix(uptime): refine host monitor checks to short-circuit TCP monitors while allowing HTTP/HTTPS checks --- backend/internal/services/uptime_service.go | 24 +- .../internal/services/uptime_service_test.go | 271 ++++++ docs/plans/current_spec.md | 898 ++++++------------ 3 files changed, 597 insertions(+), 596 deletions(-) diff --git a/backend/internal/services/uptime_service.go b/backend/internal/services/uptime_service.go index 8ecc6d4b..68c5628b 100644 --- a/backend/internal/services/uptime_service.go +++ b/backend/internal/services/uptime_service.go @@ -373,12 +373,32 @@ func (s *UptimeService) CheckAll() { // Check each host's monitors for hostID, monitors := range hostMonitors { - // If host is down, mark all monitors as down without individual checks + // If host is down, only short-circuit TCP monitors. + // HTTP/HTTPS monitors remain URL-truth authoritative and must still run checkMonitor. if hostID != "" { var uptimeHost models.UptimeHost if err := s.DB.Where("id = ?", hostID).First(&uptimeHost).Error; err == nil { if uptimeHost.Status == "down" { - s.markHostMonitorsDown(monitors, &uptimeHost) + tcpMonitors := make([]models.UptimeMonitor, 0, len(monitors)) + nonTCPMonitors := make([]models.UptimeMonitor, 0, len(monitors)) + + for _, monitor := range monitors { + normalizedType := strings.ToLower(strings.TrimSpace(monitor.Type)) + if normalizedType == "tcp" { + tcpMonitors = append(tcpMonitors, monitor) + continue + } + nonTCPMonitors = append(nonTCPMonitors, monitor) + } + + if len(tcpMonitors) > 0 { + s.markHostMonitorsDown(tcpMonitors, &uptimeHost) + } + + for _, monitor := range nonTCPMonitors { + go s.checkMonitor(monitor) + } + continue } } diff --git a/backend/internal/services/uptime_service_test.go b/backend/internal/services/uptime_service_test.go index d9fc526a..e5480ce1 100644 --- a/backend/internal/services/uptime_service_test.go +++ b/backend/internal/services/uptime_service_test.go @@ -820,6 +820,277 @@ func TestUptimeService_CheckAll_Errors(t *testing.T) { }) } +func TestUptimeService_CheckAll_HostDown_PartitionsByMonitorType(t *testing.T) { + db := setupUptimeTestDB(t) + ns := NewNotificationService(db) + us := newTestUptimeService(t, db, ns) + + us.config.TCPTimeout = 50 * time.Millisecond + us.config.MaxRetries = 0 + us.config.FailureThreshold = 1 + us.config.CheckTimeout = 2 * time.Second + + listener, err := net.Listen("tcp", "127.0.0.1:0") + assert.NoError(t, err) + addr := listener.Addr().(*net.TCPAddr) + + server := &http.Server{ + Handler: http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + w.WriteHeader(http.StatusOK) + }), + ReadHeaderTimeout: 10 * time.Second, + } + go func() { _ = server.Serve(listener) }() + t.Cleanup(func() { + _ = server.Close() + _ = listener.Close() + }) + + closedListener, err := net.Listen("tcp", "127.0.0.1:0") + assert.NoError(t, err) + closedPort := closedListener.Addr().(*net.TCPAddr).Port + _ = closedListener.Close() + + uptimeHost := models.UptimeHost{ + Host: "127.0.0.2", + Name: "Down Host", + Status: "pending", + } + err = db.Create(&uptimeHost).Error + assert.NoError(t, err) + + hostID := uptimeHost.ID + httpMonitor := models.UptimeMonitor{ + ID: "hostdown-http-monitor", + Name: "HTTP Monitor", + Type: "http", + URL: fmt.Sprintf("http://127.0.0.1:%d", addr.Port), + Enabled: true, + Status: "pending", + UptimeHostID: &hostID, + MaxRetries: 1, + } + tcpMonitor := models.UptimeMonitor{ + ID: "hostdown-tcp-monitor", + Name: "TCP Monitor", + Type: "tcp", + URL: fmt.Sprintf("127.0.0.2:%d", closedPort), + Enabled: true, + Status: "up", + UptimeHostID: &hostID, + MaxRetries: 1, + } + err = db.Create(&httpMonitor).Error + assert.NoError(t, err) + err = db.Create(&tcpMonitor).Error + assert.NoError(t, err) + + us.CheckAll() + + assert.Eventually(t, func() bool { + var refreshed models.UptimeHost + if db.Where("id = ?", uptimeHost.ID).First(&refreshed).Error != nil { + return false + } + return refreshed.Status == "down" + }, 3*time.Second, 25*time.Millisecond) + + assert.Eventually(t, func() bool { + var refreshed models.UptimeMonitor + if db.Where("id = ?", httpMonitor.ID).First(&refreshed).Error != nil { + return false + } + return refreshed.Status == "up" + }, 3*time.Second, 25*time.Millisecond) + + assert.Eventually(t, func() bool { + var refreshed models.UptimeMonitor + if db.Where("id = ?", tcpMonitor.ID).First(&refreshed).Error != nil { + return false + } + return refreshed.Status == "down" + }, 3*time.Second, 25*time.Millisecond) + + var httpHeartbeat models.UptimeHeartbeat + err = db.Where("monitor_id = ?", httpMonitor.ID).Order("created_at desc").First(&httpHeartbeat).Error + assert.NoError(t, err) + assert.Equal(t, "up", httpHeartbeat.Status) + assert.Contains(t, httpHeartbeat.Message, "HTTP 200") + assert.NotContains(t, httpHeartbeat.Message, "Host unreachable") + + var tcpHeartbeat models.UptimeHeartbeat + err = db.Where("monitor_id = ?", tcpMonitor.ID).Order("created_at desc").First(&tcpHeartbeat).Error + assert.NoError(t, err) + assert.Equal(t, "down", tcpHeartbeat.Status) + assert.Equal(t, "Host unreachable", tcpHeartbeat.Message) +} + +func TestUptimeService_CheckAll_ManualScheduledParity_ForHTTPOnHostDown(t *testing.T) { + db := setupUptimeTestDB(t) + ns := NewNotificationService(db) + us := newTestUptimeService(t, db, ns) + + us.config.TCPTimeout = 50 * time.Millisecond + us.config.MaxRetries = 0 + us.config.FailureThreshold = 1 + us.config.CheckTimeout = 2 * time.Second + + listener, err := net.Listen("tcp", "127.0.0.1:0") + assert.NoError(t, err) + addr := listener.Addr().(*net.TCPAddr) + + server := &http.Server{ + Handler: http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + w.WriteHeader(http.StatusOK) + }), + ReadHeaderTimeout: 10 * time.Second, + } + go func() { _ = server.Serve(listener) }() + t.Cleanup(func() { + _ = server.Close() + _ = listener.Close() + }) + + uptimeHost := models.UptimeHost{ + Host: "127.0.0.2", + Name: "Parity Host", + Status: "pending", + } + err = db.Create(&uptimeHost).Error + assert.NoError(t, err) + + hostID := uptimeHost.ID + manualMonitor := models.UptimeMonitor{ + ID: "manual-http-parity", + Name: "Manual HTTP", + Type: "http", + URL: fmt.Sprintf("http://127.0.0.1:%d", addr.Port), + Enabled: true, + Status: "pending", + UptimeHostID: &hostID, + MaxRetries: 1, + } + scheduledMonitor := models.UptimeMonitor{ + ID: "scheduled-http-parity", + Name: "Scheduled HTTP", + Type: "http", + URL: fmt.Sprintf("http://127.0.0.1:%d", addr.Port), + Enabled: true, + Status: "pending", + UptimeHostID: &hostID, + MaxRetries: 1, + } + err = db.Create(&manualMonitor).Error + assert.NoError(t, err) + err = db.Create(&scheduledMonitor).Error + assert.NoError(t, err) + + us.CheckMonitor(manualMonitor) + + assert.Eventually(t, func() bool { + var refreshed models.UptimeMonitor + if db.Where("id = ?", manualMonitor.ID).First(&refreshed).Error != nil { + return false + } + return refreshed.Status == "up" + }, 2*time.Second, 25*time.Millisecond) + + us.CheckAll() + + assert.Eventually(t, func() bool { + var refreshed models.UptimeMonitor + if db.Where("id = ?", scheduledMonitor.ID).First(&refreshed).Error != nil { + return false + } + return refreshed.Status == "up" + }, 3*time.Second, 25*time.Millisecond) + + var manualResult models.UptimeMonitor + err = db.Where("id = ?", manualMonitor.ID).First(&manualResult).Error + assert.NoError(t, err) + + var scheduledResult models.UptimeMonitor + err = db.Where("id = ?", scheduledMonitor.ID).First(&scheduledResult).Error + assert.NoError(t, err) + + assert.Equal(t, "up", manualResult.Status) + assert.Equal(t, manualResult.Status, scheduledResult.Status) +} + +func TestUptimeService_CheckAll_ReachableHost_StillUsesHTTPResult(t *testing.T) { + db := setupUptimeTestDB(t) + ns := NewNotificationService(db) + us := newTestUptimeService(t, db, ns) + + us.config.TCPTimeout = 50 * time.Millisecond + us.config.MaxRetries = 0 + us.config.FailureThreshold = 1 + us.config.CheckTimeout = 2 * time.Second + + listener, err := net.Listen("tcp", "127.0.0.1:0") + assert.NoError(t, err) + addr := listener.Addr().(*net.TCPAddr) + + server := &http.Server{ + Handler: http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + w.WriteHeader(http.StatusInternalServerError) + }), + ReadHeaderTimeout: 10 * time.Second, + } + go func() { _ = server.Serve(listener) }() + t.Cleanup(func() { + _ = server.Close() + _ = listener.Close() + }) + + uptimeHost := models.UptimeHost{ + Host: "127.0.0.1", + Name: "Reachable Host", + Status: "pending", + } + err = db.Create(&uptimeHost).Error + assert.NoError(t, err) + + hostID := uptimeHost.ID + httpMonitor := models.UptimeMonitor{ + ID: "reachable-host-http-fail", + Name: "Reachable Host HTTP Failure", + Type: "http", + URL: fmt.Sprintf("http://127.0.0.1:%d", addr.Port), + Enabled: true, + Status: "pending", + UptimeHostID: &hostID, + MaxRetries: 1, + } + err = db.Create(&httpMonitor).Error + assert.NoError(t, err) + + us.CheckAll() + + assert.Eventually(t, func() bool { + var refreshedHost models.UptimeHost + if db.Where("id = ?", uptimeHost.ID).First(&refreshedHost).Error != nil { + return false + } + return refreshedHost.Status == "up" + }, 3*time.Second, 25*time.Millisecond) + + assert.Eventually(t, func() bool { + var refreshed models.UptimeMonitor + if db.Where("id = ?", httpMonitor.ID).First(&refreshed).Error != nil { + return false + } + return refreshed.Status == "down" + }, 3*time.Second, 25*time.Millisecond) + + var heartbeat models.UptimeHeartbeat + err = db.Where("monitor_id = ?", httpMonitor.ID).Order("created_at desc").First(&heartbeat).Error + assert.NoError(t, err) + assert.Equal(t, "down", heartbeat.Status) + assert.Contains(t, heartbeat.Message, "HTTP 500") + assert.NotContains(t, heartbeat.Message, "Host unreachable") +} + func TestUptimeService_CheckMonitor_EdgeCases(t *testing.T) { t.Run("invalid URL format", func(t *testing.T) { db := setupUptimeTestDB(t) diff --git a/docs/plans/current_spec.md b/docs/plans/current_spec.md index 40be9842..a69a91c1 100644 --- a/docs/plans/current_spec.md +++ b/docs/plans/current_spec.md @@ -1,652 +1,362 @@ -# Uptime Monitoring Bug Triage & Fix Plan +# Uptime Monitoring Regression Investigation (Scheduled vs Manual) -## 1. Introduction +Date: 2026-03-01 +Owner: Planning Agent +Status: Investigation Complete, Fix Plan Proposed +Severity: High (false DOWN states on automated monitoring) -### Overview +## 1. Executive Summary -Uptime Monitoring in Charon uses a two-level check system: host-level TCP pre-checks followed by per-monitor HTTP/TCP checks. Newly added proxy hosts (specifically Wizarr and Charon itself) display as "DOWN" in the UI even though the underlying services are fully accessible. Manual refresh via the health check button on the Uptime page correctly shows "UP", but the automated background checker fails to produce the same result. +Two services (Wizarr and Charon) can flip to `DOWN` during scheduled cycles while manual checks immediately return `UP` because scheduled checks use a host-level TCP gate that can short-circuit monitor-level HTTP checks. -### Objectives +The scheduled path is: +- `ticker -> CheckAll -> checkAllHosts -> (host status down) -> markHostMonitorsDown` -1. Eliminate false "DOWN" status for newly added proxy hosts -2. Ensure the background checker produces consistent results with manual health checks -3. Improve the initial monitor lifecycle (creation → first check → display) -4. Address the dual `UptimeService` instance functional inconsistency -5. Evaluate whether a "custom health endpoint URL" feature is warranted +The manual path is: +- `POST /api/v1/uptime/monitors/:id/check -> CheckMonitor -> checkMonitor` -### Scope +Only the scheduled path runs host precheck gating. If host precheck fails (TCP to upstream host/port), `CheckAll` skips HTTP checks and forcibly writes monitor status to `down` with heartbeat message `Host unreachable`. -- **Backend**: `backend/internal/services/uptime_service.go`, `backend/internal/api/routes/routes.go`, `backend/internal/api/handlers/proxy_host_handler.go` -- **Frontend**: `frontend/src/pages/Uptime.tsx`, `frontend/src/api/uptime.ts` -- **Models**: `backend/internal/models/uptime.go`, `backend/internal/models/uptime_host.go` -- **Tests**: `backend/internal/services/uptime_service_test.go` (1519 LOC), `uptime_service_unit_test.go` (257 LOC), `uptime_service_race_test.go` (402 LOC), `tests/monitoring/uptime-monitoring.spec.ts` (E2E) +This is a backend state mutation problem (not only UI rendering). ---- +## 1.1 Monitoring Policy (Authoritative Behavior) + +Charon uptime monitoring SHALL follow URL-truth semantics for HTTP/HTTPS monitors, +matching third-party external monitor behavior (Uptime Kuma style) without requiring +any additional service. + +Policy: +- HTTP/HTTPS monitors are URL-truth based. The monitor result is authoritative based + on the configured URL check outcome (status code/timeout/TLS/connectivity from URL + perspective). +- Internal TCP reachability precheck (`ForwardHost:ForwardPort`) is + non-authoritative for HTTP/HTTPS monitor status. +- TCP monitors remain endpoint-socket checks and may rely on direct socket + reachability semantics. +- Host precheck may still be used for optimization, grouping telemetry, and operator + diagnostics, but SHALL NOT force HTTP/HTTPS monitors to DOWN. ## 2. Research Findings -### 2.1 Root Cause #1: Port Mismatch in Host-Level TCP Check (FIXED) +### 2.1 Execution Path Comparison (Required) -**Status**: Fixed in commit `209b2fc8`, refactored in `bfc19ef3`. +### Scheduled path behavior +- Entry: `backend/internal/api/routes/routes.go` (background ticker, calls `uptimeService.CheckAll()`) +- `CheckAll()` calls `checkAllHosts()` first. + - File: `backend/internal/services/uptime_service.go:354` +- `checkAllHosts()` updates each `UptimeHost.Status` via TCP checks in `checkHost()`. + - File: `backend/internal/services/uptime_service.go:395` +- `checkHost()` dials `UptimeHost.Host` + monitor port (prefer `ProxyHost.ForwardPort`, fallback to URL port). + - File: `backend/internal/services/uptime_service.go:437` +- Back in `CheckAll()`, monitors are grouped by `UptimeHostID`. + - File: `backend/internal/services/uptime_service.go:367` +- If `UptimeHost.Status == "down"`, `markHostMonitorsDown()` is called and individual monitor checks are skipped. + - File: `backend/internal/services/uptime_service.go:381` + - File: `backend/internal/services/uptime_service.go:593` + +### Manual path behavior +- Entry: `POST /api/v1/uptime/monitors/:id/check`. + - Handler: `backend/internal/api/handlers/uptime_handler.go:107` +- Calls `service.CheckMonitor(*monitor)` asynchronously. + - File: `backend/internal/services/uptime_service.go:707` +- `checkMonitor()` performs direct HTTP/TCP monitor check and updates monitor + heartbeat. + - File: `backend/internal/services/uptime_service.go:711` + +### Key divergence +- Scheduled: host-gated (precheck can override monitor) +- Manual: direct monitor check (no host gate) -The `checkHost()` function extracted the port from the monitor's public URL (e.g., 443 for HTTPS) instead of using `ProxyHost.ForwardPort` (e.g., 5690 for Wizarr). This caused TCP checks to fail, marking the host as `down`, which then skipped individual HTTP monitor checks. +## 3. Root Cause With Evidence -**Fix applied**: Added `Preload("ProxyHost")` and prioritized `monitor.ProxyHost.ForwardPort` over `extractPort(monitor.URL)`. +## 3.1 Primary Root Cause: Host Precheck Overrides HTTP Success in Scheduled Cycles + +When `UptimeHost` is marked `down`, scheduled checks do not run `checkMonitor()` for that host's monitors. Instead they call `markHostMonitorsDown()` which: +- sets each monitor `Status = "down"` +- writes `UptimeHeartbeat{Status: "down", Message: "Host unreachable"}` +- maxes failure count (`FailureCount = MaxRetries`) + +Evidence: +- Short-circuit: `backend/internal/services/uptime_service.go:381` +- Forced down write: `backend/internal/services/uptime_service.go:610` +- Forced heartbeat message: `backend/internal/services/uptime_service.go:624` + +This exactly matches symptom pattern: +1. Manual refresh sets monitor `UP` via direct HTTP check. +2. Next scheduler cycle can force it back to `DOWN` from host precheck path. + +## 3.2 Hypothesis Check: TCP precheck can fail while public URL HTTP check succeeds + +Confirmed as plausible by design: +- `checkHost()` tests upstream reachability (`ForwardHost:ForwardPort`) from Charon runtime. +- `checkMonitor()` tests monitor URL (public domain URL, often via Caddy/public routing). + +A service can be publicly reachable by monitor URL while upstream TCP precheck fails due to network namespace/routing/DNS/hairpin differences. + +This is especially likely for: +- self-referential routes (Charon monitoring Charon via public hostname) +- host/container networking asymmetry +- services reachable through proxy path but not directly on upstream socket from current runtime context + +## 3.3 Recent Change Correlation (Required) + +### `SyncAndCheckForHost` (regression amplifier) +- Introduced in commit `2cd19d89` and called from proxy host create path. +- Files: + - `backend/internal/services/uptime_service.go:1195` + - `backend/internal/api/handlers/proxy_host_handler.go:418` +- Behavior: creates/syncs monitor and immediately runs `checkMonitor()`. + +Impact: makes monitors quickly show `UP` after create/manual, then scheduler can flip to `DOWN` if host precheck fails. This increased visibility of scheduled/manual inconsistency. + +### `CleanupStaleFailureCounts` +- Introduced in `2cd19d89`, refined in `7a12ab79`. +- File: `backend/internal/services/uptime_service.go:1277` +- It runs at startup and resets stale monitor states only; not per-cycle override logic. +- Not root cause of recurring per-cycle flip. + +### Frontend effective status changes +- Latest commit `0241de69` refactors `effectiveStatus` handling. +- File: `frontend/src/pages/Uptime.tsx`. +- Backend evidence proves this is not visual-only: scheduler writes `down` heartbeats/messages directly in DB. + +## 3.4 Grouping Logic Analysis (`UptimeHost`/`UpstreamHost`) + +Monitors are grouped by `UptimeHostID` in `CheckAll()`. `UptimeHost` is derived from `ProxyHost.ForwardHost` in sync flows. + +Relevant code: +- group map by `UptimeHostID`: `backend/internal/services/uptime_service.go:367` +- host linkage in sync: `backend/internal/services/uptime_service.go:189`, `backend/internal/services/uptime_service.go:226` +- sync single-host update path: `backend/internal/services/uptime_service.go:1023` + +Risk: one host precheck failure can mark all grouped monitors down without URL-level validation. + +## 4. Technical Specification (Fix Plan) + +## 4.1 Minimal Proper Fix (First) + +Goal: eliminate false DOWN while preserving existing behavior as much as possible. + +Change `CheckAll()` host-down branch to avoid hard override for HTTP/HTTPS monitors. + +Mandatory hotfix rule: +- WHEN a host precheck is `down`, THE SYSTEM SHALL partition host monitors by type inside `CheckAll()`. +- `markHostMonitorsDown` MUST be invoked only for `tcp` monitors. +- `http`/`https` monitors MUST still run through `checkMonitor()` and MUST NOT be force-written `down` by the host precheck path. +- Host precheck outcomes MAY be recorded for optimization/telemetry/grouping, but MUST NOT be treated as final status for `http`/`https` monitors. + +Proposed rule: +1. If host is down: + - For `http`/`https` monitors: still run `checkMonitor()` (do not force down). + - For `tcp` monitors: keep current host-down fast-path (`markHostMonitorsDown`) or direct tcp check. +2. If host is not down: + - Keep existing behavior (run `checkMonitor()` for all monitors). -**Evidence**: Archived in `docs/plans/archive/uptime_monitoring_diagnosis.md` and `docs/implementation/uptime_monitoring_port_fix_COMPLETE.md`. +Rationale: +- Aligns scheduled behavior with manual for URL-based monitors. +- Preserves reverse proxy product semantics where public URL availability is the source of truth. +- Minimal code delta in `CheckAll()` decision branch. +- Preserves optimization for true TCP-only monitors. -**Remaining risk**: If this fix has not been deployed to production, this remains the primary cause. If deployed, residual elevated `failure_count` values in the DB may need to be reset. +### Exact file/function targets +- `backend/internal/services/uptime_service.go` + - `CheckAll()` + - add small helper (optional): `partitionMonitorsByType(...)` -### 2.2 Root Cause #2: Dual UptimeService Instance (OPEN — Functional Inconsistency) +## 4.2 Long-Term Robust Fix (Deferred) -**File**: `backend/internal/api/routes/routes.go` +Introduce host precheck as advisory signal, not authoritative override. -Two separate `UptimeService` instances are created: +Design: +1. Add `HostReachability` result to run context (not persisted as forced monitor status). +2. Always execute per-monitor checks, but use host precheck to: + - tune retries/backoff + - annotate failure reason + - optimize notification batching +3. Optionally add feature flag: + - `feature.uptime.strict_host_precheck` (default `false`) + - allows legacy strict gating in environments that want it. -| Instance | Line | Scope | -|----------|------|-------| -| `uptimeService` | 226 | Background ticker goroutine, `ProxyHostHandler`, `/system/uptime/check` endpoint | -| `uptimeSvc` | 414 | Uptime API handler routes (List, Create, Update, Delete, Check, Sync) | +Benefits: +- Removes false DOWN caused by precheck mismatch. +- Keeps performance and batching controls. +- More explicit semantics for operators. -Both share the same `*gorm.DB` (so data consistency via DB is maintained), but each has **independent in-memory state**: +## 5. API/Schema Impact -- `pendingNotifications` map (notification batching) -- `hostMutexes` map (per-host mutex for concurrent writes) -- `batchWindow` timers +No API contract change required for minimal fix. +No database migration required for minimal fix. -**Impact**: This is a **functional inconsistency that can cause race conditions between ProxyHostHandler operations and Uptime API operations**. Specifically: - -- `ProxyHostHandler.Create()` uses instance #1 (`uptimeService`) for `SyncAndCheckForHost` -- Uptime API queries (List, GetHistory) use instance #2 (`uptimeSvc`) -- In-memory state (host mutexes, pending notifications) is **invisible between instances** - -This creates a functional bug path because: - -- When a user triggers a manual check via `POST /api/v1/uptime/monitors/:id/check`, the handler uses `uptimeSvc.CheckMonitor()`. If the monitor transitions to "down", the notification is queued in `uptimeSvc`'s `pendingNotifications` map. Meanwhile, the background checker uses `uptimeService`, which has a separate `pendingNotifications` map. -- Duplicate or missed notifications -- Independent failure debouncing state -- Mutex contention issues between the two instances - -While NOT the direct cause of the "DOWN" display bug, this is a functional inconsistency — not merely a code smell — that can produce observable bugs in notification delivery and state synchronization. - -### 2.3 Root Cause #3: No Immediate Monitor Creation on Proxy Host Create (OPEN) - -> **Note — Create ↔ Update asymmetry**: `ProxyHostHandler.Update()` already calls `SyncMonitorForHost` (established pattern). The fix for `Create` should follow the same pattern for consistency. - -When a user creates a new proxy host: - -1. The proxy host is saved to DB -2. **No uptime monitor is created** — there is no hook in `ProxyHostHandler.Create()` to trigger `SyncMonitors()` or create a monitor -3. `SyncMonitorForHost()` (called on proxy host update) only updates existing monitors — it does NOT create new ones -4. The background ticker must fire (up to 1 minute) for `SyncMonitors()` to create the monitor - -**Timeline for a new proxy host to show status**: - -- T+0s: Proxy host created via API -- T+0s to T+60s: No uptime monitor exists — Uptime page shows nothing for this host -- T+60s: Background ticker fires, `SyncMonitors()` creates monitor with `status: "pending"` -- T+60s: `CheckAll()` runs, attempts host check + individual check -- T+62s: If checks succeed, monitor `status: "up"` is saved to DB -- T+90s (worst case): Frontend polls monitors and picks up the update - -This is a poor UX experience. Users expect to see their new host on the Uptime page immediately. - -### 2.4 Root Cause #4: "pending" Status Displayed as DOWN (OPEN) - -**File**: `frontend/src/pages/Uptime.tsx`, MonitorCard component - -```tsx -const isUp = latestBeat ? latestBeat.status === 'up' : monitor.status === 'up'; -``` - -When a new monitor has `status: "pending"` and no heartbeat history: - -- `latestBeat` = `null` (no history yet) -- Falls back to `monitor.status === 'up'` -- `"pending" === "up"` → `false` -- **Displayed with red DOWN styling** - -The UI has no dedicated "pending" or "unknown" state. Between creation and first check, every monitor appears DOWN. - -### 2.5 Root Cause #5: No Initial CheckAll After Server Start Sync (OPEN) - -**File**: `backend/internal/api/routes/routes.go`, lines 455-490 - -The background goroutine flow on server start: - -1. Sleep 30 seconds -2. Call `SyncMonitors()` — creates monitors for all proxy hosts -3. **Does NOT call `CheckAll()`** -4. Start 1-minute ticker -5. First `CheckAll()` runs on first tick (~90 seconds after server start) - -This means after every server restart, all monitors sit in "pending" (displayed as DOWN) for up to 90 seconds. - -### 2.6 Concern #6: Self-Referencing Check (Charon Pinging Itself) - -If Charon has a proxy host pointing to itself (e.g., `charon.example.com` → `localhost:8080`): - -**TCP host check**: Connects to `localhost:8080` → succeeds (Gin server is running locally). - -**HTTP monitor check**: Sends GET to `https://charon.example.com` → requires DNS resolution from inside the Docker container. This may fail due to: - -- **Docker hairpin NAT**: Containers cannot reach their own published ports via the host's external IP by default -- **Split-horizon DNS**: The domain may resolve to a public IP that isn't routable from within the container -- **Caddy certificate validation**: The HTTP client might reject a self-signed or incorrectly configured cert - -When the user clicks manual refresh, the same `checkMonitor()` function runs with the same options (`WithAllowLocalhost()`, `WithMaxRedirects(0)`). If manual check succeeds but background check fails, the difference is likely **timing-dependent** — the alternating "up"/"down" pattern observed in the archived diagnosis (heartbeat records alternating between `up|HTTP 200` and `down|Host unreachable`) supports this hypothesis. - -### 2.7 Feature Gap: No Custom Health Endpoint URL - -The `UptimeMonitor` model has no `health_endpoint` or `custom_url` field. All monitors check the public root URL (`/`). This is problematic because: - -- Some services redirect root → `/login` → 302 → tracked inconsistently -- Services with dedicated health endpoints (`/health`, `/api/health`) provide more reliable status -- Self-referencing checks (Charon) could use `http://localhost:8080/api/v1/health` instead of routing through DNS/Caddy - -### 2.8 Existing Test Coverage - -| File | LOC | Focus | -|------|-----|-------| -| `uptime_service_test.go` | 1519 | Integration tests with SQLite DB | -| `uptime_service_unit_test.go` | 257 | Unit tests for service methods | -| `uptime_service_race_test.go` | 402 | Concurrency/race condition tests | -| `uptime_service_notification_test.go` | — | Notification batching tests | -| `uptime_handler_test.go` | — | Handler HTTP endpoint tests | -| `uptime_monitor_initial_state_test.go` | — | Initial state tests | -| `uptime-monitoring.spec.ts` | — | Playwright E2E (22 scenarios) | - ---- - -## 3. Technical Specifications - -### 3.1 Consolidate UptimeService Singleton - -**Current**: Two instances (`uptimeService` line 226, `uptimeSvc` line 414) in `routes.go`. - -**Target**: Single instance passed to both the background goroutine AND the API handlers. - -```go -// routes.go — BEFORE (two instances) -uptimeService := services.NewUptimeService(db, notificationService) // line 226 -uptimeSvc := services.NewUptimeService(db, notificationService) // line 414 - -// routes.go — AFTER (single instance) -uptimeService := services.NewUptimeService(db, notificationService) // line 226 -// line 414: reuse uptimeService for handler registration -uptimeHandler := handlers.NewUptimeHandler(uptimeService) -``` - -**Impact**: All in-memory state (mutexes, notification batching, pending notifications) is shared. The single instance must remain thread-safe (it already is — methods use `sync.Mutex`). - -### 3.2 Trigger Monitor Creation + Immediate Check on Proxy Host Create - -**File**: `backend/internal/api/handlers/proxy_host_handler.go` - -After successfully creating a proxy host, call `SyncMonitors()` (or a targeted sync) and trigger an immediate check: - -```go -// In Create handler, after host is saved: -if h.uptimeService != nil { - _ = h.uptimeService.SyncMonitors() - // Trigger immediate check for the new monitor - var monitor models.UptimeMonitor - if err := h.uptimeService.DB.Where("proxy_host_id = ?", host.ID).First(&monitor).Error; err == nil { - go h.uptimeService.CheckMonitor(monitor) - } -} -``` - -**Alternative (lighter-weight)**: Add a `SyncAndCheckForHost(hostID uint)` method that creates the monitor if needed and immediately checks it. - -### 3.3 Add "pending" UI State - -**File**: `frontend/src/pages/Uptime.tsx` - -Add dedicated handling for `"pending"` status: - -```tsx -const isPending = monitor.status === 'pending' && (!history || history.length === 0); -const isUp = latestBeat ? latestBeat.status === 'up' : monitor.status === 'up'; -const isPaused = monitor.enabled === false; -``` - -Visual treatment for pending state: - -- Yellow/gray pulsing indicator (distinct from DOWN red and UP green) -- Badge text: "CHECKING..." or "PENDING" -- Heartbeat bar: show empty placeholder bars with a spinner or pulse animation - -### 3.4 Run CheckAll After Initial SyncMonitors - -**File**: `backend/internal/api/routes/routes.go` - -```go -// AFTER initial sync -if enabled { - if err := uptimeService.SyncMonitors(); err != nil { - logger.Log().WithError(err).Error("Failed to sync monitors") - } - // Run initial check immediately - uptimeService.CheckAll() -} -``` - -### 3.5 Add Optional `check_url` Field to UptimeMonitor (Enhancement) - -**Model change** (`backend/internal/models/uptime.go`): - -```go -type UptimeMonitor struct { - // ... existing fields - CheckURL string `json:"check_url,omitempty" gorm:"default:null"` -} -``` - -**Service behavior** (`uptime_service.go` `checkMonitor()`): - -- If `monitor.CheckURL` is set and non-empty, use it instead of `monitor.URL` for the HTTP check -- This allows users to configure `/health` or `http://localhost:8080/api/v1/health` for self-referencing - -**Frontend**: Add an optional "Health Check URL" field in the edit monitor modal. - -**Auto-migration**: GORM handles adding the column. Existing monitors keep `CheckURL = ""` (uses default URL behavior). - -#### 3.5.1 SSRF Protection for CheckURL - -The `CheckURL` field accepts user-controlled URLs that the server will fetch. This requires layered SSRF defenses: - -**Write-time validation** (on Create/Update API): - -- Validate `CheckURL` before saving to DB -- **Scheme restriction**: Only `http://` and `https://` allowed. Block `file://`, `ftp://`, `gopher://`, and all other schemes -- **Max URL length**: 2048 characters -- Reject URLs that fail `url.Parse()` or have empty host components - -**Check-time validation** (before each HTTP request): - -- Re-validate the URL against the deny list before every check execution (defense-in-depth — the stored URL could have been valid at write time but conditions may change) -- **Localhost handling**: Allow loopback addresses (`127.0.0.1`, `::1`, `localhost`) since self-referencing checks are a valid use case. Block cloud metadata IPs: - - `169.254.169.254` (AWS/GCP/Azure instance metadata) - - `fd00::/8` (unique local addresses) - - `100.100.100.200` (Alibaba Cloud metadata) - - `169.254.0.0/16` link-local range (except loopback) -- **DNS rebinding protection**: Resolve the hostname at request time, pin the resolved IP, and validate the resolved IP against the deny list before establishing a connection. Use a custom `net.Dialer` or `http.Transport.DialContext` to enforce this -- **Redirect validation**: If `CheckURL` follows HTTP redirects (3xx), validate each redirect target URL against the same deny list (scheme, host, resolved IP). Use a `CheckRedirect` function on the `http.Client` to intercept and validate each hop - -**Implementation pattern**: - -```go -func validateCheckURL(rawURL string) error { - if len(rawURL) > 2048 { - return ErrURLTooLong - } - parsed, err := url.Parse(rawURL) - if err != nil { - return ErrInvalidURL - } - if parsed.Scheme != "http" && parsed.Scheme != "https" { - return ErrDisallowedScheme - } - if parsed.Host == "" { - return ErrEmptyHost - } - return nil -} - -func validateResolvedIP(ip net.IP) error { - // Allow loopback - if ip.IsLoopback() { - return nil - } - // Block cloud metadata and link-local - if isCloudMetadataIP(ip) || ip.IsLinkLocalUnicast() { - return ErrDeniedIP - } - return nil -} -``` - -### 3.6 Data Cleanup: Reset Stale Failure Counts - -After deploying the port fix (if not already deployed), run a one-time DB cleanup: - -```sql --- Reset failure counts for hosts/monitors stuck from the port mismatch era --- Only reset monitors with elevated failure counts AND no recent successful heartbeat -UPDATE uptime_hosts SET failure_count = 0, status = 'pending' WHERE status = 'down'; -UPDATE uptime_monitors SET failure_count = 0, status = 'pending' -WHERE status = 'down' - AND failure_count > 5 - AND id NOT IN ( - SELECT DISTINCT monitor_id FROM uptime_heartbeats - WHERE status = 'up' AND created_at > datetime('now', '-24 hours') - ); -``` - -This could be automated in `SyncMonitors()` or done via a migration. - ---- - -## 4. Data Flow Diagrams - -### Current Flow (Buggy) - -``` -[Proxy Host Created] → (no uptime action) - → [Wait up to 60s for ticker] - → SyncMonitors() creates monitor (status: "pending") - → CheckAll() runs: - → checkAllHosts() TCP to ForwardHost:ForwardPort - → If host up → checkMonitor() HTTP to public URL - → DB updated - → [Wait up to 30s for frontend poll] - → Frontend displays status -``` - -### Proposed Flow (Fixed) - -``` -[Proxy Host Created] - → SyncMonitors() or SyncAndCheckForHost() immediately - → Monitor created (status: "pending") - → Frontend shows "PENDING" (yellow indicator) - → Immediate checkMonitor() in background goroutine - → DB updated (status: "up" or "down") - → Frontend polls in 30s → shows actual status -``` - ---- - -## 5. Implementation Plan - -### Phase 1: Playwright E2E Tests (Behavior Specification) - -Define expected behavior before implementation: - -| Test | Description | -|------|-------------| -| New proxy host monitor appears immediately | After creating a proxy host, navigate to Uptime page, verify the monitor card exists | -| New monitor shows pending state | Verify "PENDING" badge before first check completes | -| Monitor status updates after check | Trigger manual check, verify status changes from pending/down to up | -| Verify no false DOWN on first load | Create host, wait for background check, verify status is UP (not DOWN) | - -**Files**: `tests/monitoring/uptime-monitoring.spec.ts` (extend existing suite) - -### Phase 2: Backend — Consolidate UptimeService Instance - -1. Remove second `NewUptimeService` call at `routes.go` line 414 -2. Pass `uptimeService` (line 226) to `NewUptimeHandler()` -3. Verify all handler operations use the shared instance -4. Update existing tests that may create multiple instances - -**Files**: `backend/internal/api/routes/routes.go` - -### Phase 3: Backend — Immediate Monitor Lifecycle - -1. In `ProxyHostHandler.Create()`, after saving host: call `SyncMonitors()` or create a targeted `SyncAndCheckForHost()` method -2. Add `CheckAll()` call after initial `SyncMonitors()` in the background goroutine -3. Consider adding a `SyncAndCheckForHost(hostID uint)` method to `UptimeService` that: - - Finds or creates the monitor for the given proxy host - - Immediately runs `checkMonitor()` in a goroutine - - Returns the monitor ID for the caller - -**Files**: `backend/internal/services/uptime_service.go`, `backend/internal/api/handlers/proxy_host_handler.go`, `backend/internal/api/routes/routes.go` - -### Phase 4: Frontend — Pending State Display - -1. Add `isPending` check in `MonitorCard` component -2. Add yellow/gray styling for pending state -3. Add pulsing animation for pending badge -4. Add i18n key `uptime.pending` → "CHECKING..." for **all 5 supported languages** (not just the default locale) -5. Ensure heartbeat bar handles zero-length history gracefully - -**Files**: `frontend/src/pages/Uptime.tsx`, `frontend/src/i18n/` locale files - -### Phase 5: Backend — Optional `check_url` Field (Enhancement) - -1. Add `CheckURL` field to `UptimeMonitor` model -2. Update `checkMonitor()` to use `CheckURL` if set -3. Update `SyncMonitors()` — do NOT overwrite user-configured `CheckURL` -4. Update API DTOs for create/update - -**Files**: `backend/internal/models/uptime.go`, `backend/internal/services/uptime_service.go`, `backend/internal/api/handlers/uptime_handler.go` - -### Phase 6: Frontend — Health Check URL in Edit Modal - -1. Add optional "Health Check URL" field to `EditMonitorModal` and `CreateMonitorModal` -2. Show placeholder text: "Leave empty to use monitor URL" -3. Validate URL format on frontend - -**Files**: `frontend/src/pages/Uptime.tsx` - -### Phase 7: Testing & Validation - -1. Run existing backend test suites (2178 LOC across 3 files) -2. Add tests for: - - Single `UptimeService` instance behavior - - Immediate monitor creation on proxy host create - - `CheckURL` fallback logic - - "pending" → "up" transition -3. Add edge case tests: - - **Rapid Create-Delete**: Proxy host created and immediately deleted before `SyncAndCheckForHost` goroutine completes — goroutine should handle non-existent proxy host gracefully (no panic, no orphaned monitor) - - **Concurrent Creates**: Multiple proxy hosts created simultaneously — verify `SyncMonitors()` from Create handlers doesn't conflict with background ticker's `SyncMonitors()` (no duplicate monitors, no data races) - - **Feature Flag Toggle**: If `feature.uptime.enabled` is toggled to `false` while immediate check goroutine is running — goroutine should exit cleanly without writing stale results - - **CheckURL with redirects**: `CheckURL` that 302-redirects to a private IP — redirect target must be validated against the deny list (SSRF redirect chain) -4. Run Playwright E2E suite with Docker rebuild -5. Verify coverage thresholds - -### Phase 8: Data Cleanup Migration - -1. Add one-time migration or startup hook to reset stale `failure_count` and `status` on hosts/monitors that were stuck from the port mismatch era -2. Log the cleanup action - ---- +Long-term fix may add one feature flag setting only. ## 6. EARS Requirements -1. WHEN a new proxy host is created, THE SYSTEM SHALL create a corresponding uptime monitor within 5 seconds (not waiting for the 1-minute ticker) -2. WHEN a new uptime monitor is created, THE SYSTEM SHALL immediately trigger a health check in a background goroutine -3. WHEN a monitor has status "pending" and no heartbeat history, THE SYSTEM SHALL display a distinct visual indicator (not DOWN red) -4. WHEN the server starts, THE SYSTEM SHALL run `CheckAll()` immediately after `SyncMonitors()` (not wait for first tick) -5. THE SYSTEM SHALL use a single `UptimeService` instance for both background checks and API handlers -6. WHERE a monitor has a `check_url` configured, THE SYSTEM SHALL use it for health checks instead of the monitor URL -7. WHEN a monitor's host-level TCP check succeeds but HTTP check fails, THE SYSTEM SHALL record the specific failure reason in the heartbeat message -8. IF the uptime feature flag is disabled, THEN THE SYSTEM SHALL skip all monitor sync and check operations +### Ubiquitous +- THE SYSTEM SHALL evaluate HTTP/HTTPS monitor availability using URL-level checks as the authoritative signal. ---- +### Event-driven +- WHEN the scheduled uptime cycle runs, THE SYSTEM SHALL execute HTTP/HTTPS monitor checks regardless of internal host precheck state. +- WHEN the scheduled uptime cycle runs and host precheck is down, THE SYSTEM SHALL apply host-level forced-down logic only to TCP monitors. -## 7. Acceptance Criteria +### State-driven +- WHILE a monitor type is `http` or `https`, THE SYSTEM SHALL NOT force monitor status to `down` solely from internal host precheck failure. +- WHILE a monitor type is `tcp`, THE SYSTEM SHALL evaluate status using endpoint socket reachability semantics. -### Must Have +### Unwanted behavior +- IF internal host precheck is unreachable AND URL-level HTTP/HTTPS check returns success, THEN THE SYSTEM SHALL set monitor status to `up`. +- IF internal host precheck is reachable AND URL-level HTTP/HTTPS check fails, THEN THE SYSTEM SHALL set monitor status to `down`. -- [ ] WHEN a new proxy host is created, a corresponding uptime monitor exists within 5 seconds -- [ ] WHEN a new uptime monitor is created, an immediate health check runs -- [ ] WHEN a monitor has status "pending", a distinct yellow/gray visual indicator is shown (not red DOWN) -- [ ] WHEN the server starts, `CheckAll()` runs immediately after `SyncMonitors()` -- [ ] Only one `UptimeService` instance exists at runtime +### Optional +- WHERE host precheck telemetry is enabled, THE SYSTEM SHALL record host-level reachability for diagnostics and grouping without overriding HTTP/HTTPS monitor final state. -### Should Have +## 7. Implementation Plan -- [ ] WHEN a monitor has a `check_url` configured, it is used for health checks -- [ ] WHEN a monitor's host-level TCP check succeeds but HTTP check fails, the heartbeat message contains the failure reason -- [ ] Stale `failure_count` values from the port mismatch era are reset on deployment +### Phase 1: Reproduction Lock-In (Tests First) +- Add backend service test proving current regression: + - host precheck fails + - monitor URL check would succeed + - scheduled `CheckAll()` currently writes down (existing behavior) +- File: `backend/internal/services/uptime_service_test.go` (new test block) -### Nice to Have +### Phase 2: Minimal Backend Fix +- Update `CheckAll()` branch logic to run HTTP/HTTPS monitors even when host is down. +- Make monitor partitioning explicit and mandatory in `CheckAll()` host-down branch. +- Add an implementation guard before partitioning: normalize monitor type using + `strings.TrimSpace` + `strings.ToLower` to prevent `HTTP`/`HTTPS` case + regressions and whitespace-related misclassification. +- Ensure `markHostMonitorsDown` is called only for TCP monitor partitions. +- File: `backend/internal/services/uptime_service.go` -- [ ] Dedicated UI indicator for "first check in progress" (animated pulse) -- [ ] Automatic detection of health endpoints (try `/health` first, fall back to `/`) +### Phase 3: Backend Validation +- Add/adjust tests: + - scheduled path no longer forces down when HTTP succeeds + - manual and scheduled reach same final state for HTTP monitors + - internal host unreachable + public URL HTTP 200 => monitor is `UP` + - internal host reachable + public URL failure => monitor is `DOWN` + - TCP monitor behavior unchanged under host-down conditions +- Files: + - `backend/internal/services/uptime_service_test.go` + - `backend/internal/services/uptime_service_race_test.go` (if needed for concurrency side-effects) ---- +### Phase 4: Integration/E2E Coverage +- Add targeted API-level integration test for scheduler vs manual parity. +- Add Playwright scenario for: + - monitor set UP by manual check + - remains UP after scheduled cycle when URL is reachable +- Add parity scenario for: + - internal TCP precheck unreachable + URL returns 200 => `UP` + - internal TCP precheck reachable + URL failure => `DOWN` +- Files: + - `backend/internal/api/routes/routes_test.go` (or uptime handler integration suite) + - `tests/monitoring/uptime-monitoring.spec.ts` (or equivalent uptime spec file) -## 8. PR Slicing Strategy +Scope note: +- This hotfix plan is intentionally limited to backend behavior correction and + regression tests (unit/integration/E2E). +- Dedicated documentation-phase work is deferred and out of scope for this + hotfix PR. -### Decision: 3 PRs +## 8. Test Plan (Unit / Integration / E2E) -**Trigger reasons**: Cross-domain changes (backend + frontend + model), independent concerns (UX fix vs backend architecture vs new feature), review size management. +Duplicate notification definition (hotfix acceptance/testing): +- A duplicate notification means the same `(monitor_id, status, + scheduler_tick_id)` is emitted more than once within a single scheduler run. -### PR-1: Backend Bug Fixes (Architecture + Lifecycle) +## Unit Tests +1. `CheckAll_HostDown_DoesNotForceDown_HTTPMonitor_WhenHTTPCheckSucceeds` +2. `CheckAll_HostDown_StillHandles_TCPMonitor_Conservatively` +3. `CheckAll_ManualAndScheduledParity_HTTPMonitor` +4. `CheckAll_InternalHostUnreachable_PublicURL200_HTTPMonitorEndsUp` (blocking) +5. `CheckAll_InternalHostReachable_PublicURLFail_HTTPMonitorEndsDown` (blocking) -**Scope**: Phases 2, 3, and initial CheckAll (Section 3.4) +## Integration Tests +1. Scheduler endpoint (`/api/v1/system/uptime/check`) parity with monitor check endpoint. +2. Verify DB heartbeat message is real HTTP result (not `Host unreachable`) for HTTP monitors where URL is reachable. +3. Verify when host precheck is down, HTTP monitor heartbeat/notification output is derived from `checkMonitor()` (not synthetic host-path `Host unreachable`). +4. Verify no duplicate notifications are emitted from host+monitor paths for the same scheduler run, where duplicate is defined as repeated `(monitor_id, status, scheduler_tick_id)`. +5. Verify internal host precheck unreachable + public URL 200 still resolves monitor `UP`. +6. Verify internal host precheck reachable + public URL failure resolves monitor `DOWN`. -**Files**: +## E2E Tests +1. Create/sync monitor scenario where manual refresh returns `UP`. +2. Wait one scheduler interval. +3. Assert monitor remains `UP` and latest heartbeat is not forced `Host unreachable` for reachable URL. +4. Assert scenario: internal host precheck unreachable + public URL 200 => monitor remains `UP`. +5. Assert scenario: internal host precheck reachable + public URL failure => monitor is `DOWN`. -- `backend/internal/api/routes/routes.go` — consolidate to single UptimeService instance, add CheckAll after initial sync -- `backend/internal/services/uptime_service.go` — add `SyncAndCheckForHost()` method -- `backend/internal/api/handlers/proxy_host_handler.go` — call SyncAndCheckForHost on Create -- Backend test files — update for single instance, add new lifecycle tests -- Data cleanup migration -- `ARCHITECTURE.md` — update to reflect the UptimeService singleton consolidation (architecture change) +## Regression Guardrails +- Add a test explicitly asserting that host precheck must not unconditionally override HTTP monitor checks. +- Add explicit assertions that HTTP monitors under host-down precheck emit + check-derived heartbeat messages and do not produce duplicate notifications + under the `(monitor_id, status, scheduler_tick_id)` rule within a single + scheduler run. -**Dependencies**: None (independent of frontend changes) +## 9. Risks and Rollback -**Validation**: All backend tests pass, no duplicate UptimeService instantiation, new proxy hosts get immediate monitors, ARCHITECTURE.md reflects current design +## Risks +1. More HTTP checks under true host outage may increase check volume. +2. Notification patterns may shift from single host-level event to monitor-level batched events. +3. Edge cases for mixed-type monitor groups (HTTP + TCP) need deterministic behavior. -**Rollback**: Revert commit; behavior returns to previous (ticker-based) lifecycle +## Mitigations +1. Preserve batching (`queueDownNotification`) and existing retry thresholds. +2. Keep TCP strict path unchanged in minimal fix. +3. Add explicit log fields and targeted tests for mixed groups. -### PR-2: Frontend Pending State +## Rollback Plan +1. Revert the `CheckAll()` branch change only (single-file rollback). +2. Keep added tests; mark expected behavior as legacy if temporary rollback needed. +3. If necessary, introduce temporary feature toggle to switch between strict and tolerant host gating. -**Scope**: Phase 4 +## 10. PR Slicing Strategy -**Files**: +Decision: Single focused PR (hotfix + tests) -- `frontend/src/pages/Uptime.tsx` — add pending state handling -- `frontend/src/i18n/` locale files — add `uptime.pending` key -- `frontend/src/pages/__tests__/Uptime.spec.tsx` — update tests +Trigger reasons: +- High-severity runtime behavior fix requiring minimal blast radius +- Fast review/rollback with behavior-only delta plus regression coverage +- Avoid scope creep into optional hardening/feature-flag work -**Dependencies**: Works independently of PR-1 (pending state display improves UX regardless of backend fix timing) +### PR-1 (Hotfix + Tests) +Scope: +- `CheckAll()` host-down branch adjustment for HTTP/HTTPS +- Unit/integration/E2E regression tests for URL-truth semantics -**Validation**: Playwright E2E tests pass, pending monitors show yellow indicator +Files: +- `backend/internal/services/uptime_service.go` +- `backend/internal/services/uptime_service_test.go` +- `backend/internal/api/routes/routes_test.go` (or equivalent) +- `tests/monitoring/uptime-monitoring.spec.ts` (or equivalent) -**Rollback**: Revert commit; pending monitors display as DOWN (existing behavior) +Validation gates: +- backend unit tests pass +- targeted uptime integration tests pass +- targeted uptime E2E tests pass +- no behavior regression in existing `CheckAll` tests -### PR-3: Custom Health Check URL (Enhancement) +Rollback: +- single revert of PR-1 commit -**Scope**: Phases 5, 6 +## 11. Acceptance Criteria (DoD) -**Files**: +1. Scheduled and manual checks produce consistent status for HTTP/HTTPS monitors. +2. A reachable monitor URL is not forced to `DOWN` solely by host precheck failure. +3. New regression tests fail before fix and pass after fix. +4. No break in TCP monitor behavior expectations. +5. No new critical/high security findings in touched paths. +6. Blocking parity case passes: internal host precheck unreachable + public URL 200 => scheduled result is `UP`. +7. Blocking parity case passes: internal host precheck reachable + public URL failure => scheduled result is `DOWN`. +8. Under host-down precheck, HTTP monitors produce check-derived heartbeat messages (not synthetic `Host unreachable` from host path). +9. No duplicate notifications are produced by host+monitor paths within a + single scheduler run, where duplicate is defined as repeated + `(monitor_id, status, scheduler_tick_id)`. -- `backend/internal/models/uptime.go` — add CheckURL field -- `backend/internal/services/uptime_service.go` — use CheckURL in checkMonitor -- `backend/internal/api/handlers/uptime_handler.go` — update DTOs -- `frontend/src/pages/Uptime.tsx` — add form field -- Test files — add coverage for CheckURL logic +## 12. Implementation Risks -**Dependencies**: PR-1 should be merged first (shared instance simplifies testing) +1. Increased scheduler workload during host-precheck failures because HTTP/HTTPS checks continue to run. +2. Notification cadence may change due to check-derived monitor outcomes replacing host-forced synthetic downs. +3. Mixed monitor groups (TCP + HTTP/HTTPS) require strict ordering/partitioning to avoid regression. -**Validation**: Create monitor with custom health URL, verify check uses it - -**Rollback**: Revert commit; GORM auto-migration adds the column but it remains unused - ---- - -## 9. Risk Assessment - -| Risk | Severity | Likelihood | Mitigation | -|------|----------|------------|------------| -| Consolidating UptimeService instance introduces race conditions | High | Low | Existing mutex protections are designed for shared use; run race tests with `-race` flag | -| Immediate SyncMonitors on proxy host create adds latency to API response | Medium | Medium | Run SyncAndCheckForHost in a goroutine; return HTTP 201 immediately | -| "pending" UI state confuses users who expect UP/DOWN binary | Low | Low | Clear tooltip/label: "Initial health check in progress..." | -| CheckURL allows SSRF if user provides malicious URL | High | Low | Layered SSRF defense (see Section 3.5.1): write-time validation (scheme, length, parse), check-time re-validation, DNS rebinding protection (pin resolved IP against deny list), redirect chain validation. Allow loopback for self-referencing checks; block cloud metadata IPs (`169.254.169.254`, `fd00::`, etc.) | -| Data cleanup migration resets legitimate DOWN status | Medium | Medium | Only reset monitors with elevated failure counts AND no recent successful heartbeat | -| Self-referencing check (Charon) still fails due to Docker DNS | Medium | High | **PR-3 scope**: When `SyncMonitors()` creates a monitor, if `ForwardHost` resolves to loopback (`localhost`, `127.0.0.1`, or the container's own hostname), automatically set `CheckURL` to `http://{ForwardHost}:{ForwardPort}/` to bypass the DNS/Caddy round-trip. Tracked as technical debt if deferred beyond PR-3 | - ---- - -## 10. Validation Plan (Mandatory Sequence) - -0. **E2E environment prerequisite** - - Determine rebuild necessity per testing policy: if application/runtime or Docker input changes are present, rebuild is required. - - If rebuild is required or the container is unhealthy, run `.github/skills/scripts/skill-runner.sh docker-rebuild-e2e`. - - Record container health outcome before executing tests. - -1. **Playwright first** - - Run targeted uptime monitoring E2E scenarios. - -2. **Local patch coverage preflight** - - Generate `test-results/local-patch-report.md` and `test-results/local-patch-report.json`. - -3. **Unit and coverage** - - Backend coverage run (threshold >= 85%). - - Frontend coverage run (threshold >= 85%). - -4. **Race condition tests** - - Run `go test -race ./backend/internal/services/...` to verify single-instance thread safety. - -5. **Type checks** - - Frontend TypeScript check. - -6. **Pre-commit** - - `pre-commit run --all-files` with zero blocking failures. - -7. **Security scans** - - CodeQL Go + JS (security-and-quality). - - GORM security scan (model changes in PR-3). - - Trivy scan. - -8. **Build verification** - - Backend build + frontend build pass. - ---- - -## 11. Architecture Reference - -### Two-Level Check System - -``` -Level 1: Host-Level TCP Pre-Check -├── Purpose: Quickly determine if backend host/container is reachable -├── Method: TCP connection to ForwardHost:ForwardPort -├── Runs: Once per unique UptimeHost -├── If DOWN → Skip all Level 2 checks, mark all monitors DOWN -└── If UP → Proceed to Level 2 - -Level 2: Service-Level HTTP/TCP Check -├── Purpose: Verify specific service is responding correctly -├── Method: HTTP GET to monitor URL (or CheckURL if set) -├── Runs: Per-monitor (in parallel goroutines) -└── Accepts: 2xx, 3xx, 401, 403 as "up" -``` - -### Background Ticker Flow - -``` -Server Start → Sleep 30s → SyncMonitors() - → [PROPOSED] CheckAll() - → Start 1-minute ticker - → Each tick: SyncMonitors() → CheckAll() - → checkAllHosts() [parallel, staggered] - → Group monitors by host - → For each host: - If down → markHostMonitorsDown() - If up → checkMonitor() per monitor [parallel goroutines] -``` - -### Key Configuration Values - -| Setting | Value | Source | -|---------|-------|--------| -| `batchWindow` | 30s | `NewUptimeService()` | -| `TCPTimeout` | 10s | `NewUptimeService()` | -| `MaxRetries` (host) | 2 | `NewUptimeService()` | -| `FailureThreshold` (host) | 2 | `NewUptimeService()` | -| `CheckTimeout` | 60s | `NewUptimeService()` | -| `StaggerDelay` | 100ms | `NewUptimeService()` | -| `MaxRetries` (monitor) | 3 | `UptimeMonitor.MaxRetries` default | -| Ticker interval | 1 min | `routes.go` ticker | -| Frontend poll interval | 30s | `Uptime.tsx` refetchInterval | -| History poll interval | 60s | `MonitorCard` refetchInterval | - ---- - -## 12. Rollback and Contingency - -1. **PR-1**: If consolidating UptimeService causes regressions → revert commit; background checker and API revert to two separate instances (existing behavior). -2. **PR-2**: If pending state display causes confusion → revert commit; monitors display DOWN for pending (existing behavior). -3. **PR-3**: If CheckURL introduces SSRF or regressions → revert commit; column stays in DB but is unused. -4. **Data cleanup**: If migration resets legitimate DOWN hosts → restore from SQLite backup (standard Charon backup flow). - -Post-rollback smoke checks: -- Verify background ticker creates monitors for all proxy hosts -- Verify manual health check button produces correct status -- Verify notification batching works correctly +Mitigations: +- Keep change localized to `CheckAll()` host-down branch decisioning. +- Add explicit regression tests for both parity directions and mixed monitor types. +- Keep rollback path as single-commit revert. From fbb86b1cc3338eb69d1c552e9b10da4b719c219c Mon Sep 17 00:00:00 2001 From: "renovate[bot]" <29139614+renovate[bot]@users.noreply.github.com> Date: Mon, 2 Mar 2026 03:15:19 +0000 Subject: [PATCH 37/38] chore(deps): update non-major-updates --- .github/workflows/renovate.yml | 2 +- .github/workflows/security-pr.yml | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/.github/workflows/renovate.yml b/.github/workflows/renovate.yml index dd73e2cd..6d17aa86 100644 --- a/.github/workflows/renovate.yml +++ b/.github/workflows/renovate.yml @@ -25,7 +25,7 @@ jobs: fetch-depth: 1 - name: Run Renovate - uses: renovatebot/github-action@8d75b92f43899d483728e9a8a7fd44238020f6e6 # v46.1.2 + uses: renovatebot/github-action@7b4b65bf31e07d4e3e51708d07700fb41bc03166 # v46.1.3 with: configurationFile: .github/renovate.json token: ${{ secrets.RENOVATE_TOKEN || secrets.GITHUB_TOKEN }} diff --git a/.github/workflows/security-pr.yml b/.github/workflows/security-pr.yml index c02e9da2..8eeb9569 100644 --- a/.github/workflows/security-pr.yml +++ b/.github/workflows/security-pr.yml @@ -362,7 +362,7 @@ jobs: - name: Run Trivy filesystem scan (SARIF output) if: steps.check-artifact.outputs.artifact_exists == 'true' || github.event_name == 'push' || github.event_name == 'pull_request' # aquasecurity/trivy-action v0.33.1 - uses: aquasecurity/trivy-action@1bd062560b422f5944df1de50abd05162bea079e + uses: aquasecurity/trivy-action@4c61e6329bab9be735ca35291551614bc663dff3 with: scan-type: 'fs' scan-ref: ${{ steps.extract.outputs.binary_path }} @@ -394,7 +394,7 @@ jobs: - name: Run Trivy filesystem scan (fail on CRITICAL/HIGH) if: steps.check-artifact.outputs.artifact_exists == 'true' || github.event_name == 'push' || github.event_name == 'pull_request' # aquasecurity/trivy-action v0.33.1 - uses: aquasecurity/trivy-action@1bd062560b422f5944df1de50abd05162bea079e + uses: aquasecurity/trivy-action@4c61e6329bab9be735ca35291551614bc663dff3 with: scan-type: 'fs' scan-ref: ${{ steps.extract.outputs.binary_path }} From 10259146df8f3d706f6cf4a916e55e8c20be5bfd Mon Sep 17 00:00:00 2001 From: GitHub Actions Date: Mon, 2 Mar 2026 03:40:08 +0000 Subject: [PATCH 38/38] fix(uptime): implement initial uptime bootstrap logic and related tests --- .../api/handlers/proxy_host_handler_test.go | 57 ++++++++++ backend/internal/api/routes/routes.go | 41 +++++-- .../api/routes/routes_coverage_test.go | 52 +++++++++ .../routes/routes_uptime_bootstrap_test.go | 107 ++++++++++++++++++ .../services/uptime_service_pr1_test.go | 71 ++++++++++++ frontend/src/pages/__tests__/Uptime.test.tsx | 17 +++ 6 files changed, 333 insertions(+), 12 deletions(-) create mode 100644 backend/internal/api/routes/routes_uptime_bootstrap_test.go diff --git a/backend/internal/api/handlers/proxy_host_handler_test.go b/backend/internal/api/handlers/proxy_host_handler_test.go index 022f1141..cb2f984f 100644 --- a/backend/internal/api/handlers/proxy_host_handler_test.go +++ b/backend/internal/api/handlers/proxy_host_handler_test.go @@ -9,6 +9,7 @@ import ( "net/http/httptest" "strings" "testing" + "time" "github.com/gin-gonic/gin" "github.com/google/uuid" @@ -68,6 +69,33 @@ func setupTestRouterWithReferenceTables(t *testing.T) (*gin.Engine, *gorm.DB) { return r, db } +func setupTestRouterWithUptime(t *testing.T) (*gin.Engine, *gorm.DB) { + t.Helper() + + dsn := "file:" + t.Name() + "?mode=memory&cache=shared" + db, err := gorm.Open(sqlite.Open(dsn), &gorm.Config{}) + require.NoError(t, err) + require.NoError(t, db.AutoMigrate( + &models.ProxyHost{}, + &models.Location{}, + &models.Notification{}, + &models.NotificationProvider{}, + &models.UptimeMonitor{}, + &models.UptimeHeartbeat{}, + &models.UptimeHost{}, + &models.Setting{}, + )) + + ns := services.NewNotificationService(db) + us := services.NewUptimeService(db, ns) + h := NewProxyHostHandler(db, nil, ns, us) + r := gin.New() + api := r.Group("/api/v1") + h.RegisterRoutes(api) + + return r, db +} + func TestProxyHostHandler_ResolveAccessListReference_TargetedBranches(t *testing.T) { t.Parallel() @@ -201,6 +229,35 @@ func TestProxyHostCreate_ReferenceResolution_TargetedBranches(t *testing.T) { }) } +func TestProxyHostCreate_TriggersAsyncUptimeSyncWhenServiceConfigured(t *testing.T) { + t.Parallel() + + router, db := setupTestRouterWithUptime(t) + + upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + w.WriteHeader(http.StatusOK) + })) + t.Cleanup(upstream.Close) + + domain := strings.TrimPrefix(upstream.URL, "http://") + body := fmt.Sprintf(`{"name":"Uptime Hook","domain_names":"%s","forward_scheme":"http","forward_host":"app-service","forward_port":8080,"enabled":true}`, domain) + req := httptest.NewRequest(http.MethodPost, "/api/v1/proxy-hosts", strings.NewReader(body)) + req.Header.Set("Content-Type", "application/json") + + resp := httptest.NewRecorder() + router.ServeHTTP(resp, req) + require.Equal(t, http.StatusCreated, resp.Code) + + var created models.ProxyHost + require.NoError(t, db.Where("domain_names = ?", domain).First(&created).Error) + + var count int64 + require.Eventually(t, func() bool { + db.Model(&models.UptimeMonitor{}).Where("proxy_host_id = ?", created.ID).Count(&count) + return count > 0 + }, 3*time.Second, 50*time.Millisecond) +} + func TestProxyHostLifecycle(t *testing.T) { t.Parallel() router, _ := setupTestRouter(t) diff --git a/backend/internal/api/routes/routes.go b/backend/internal/api/routes/routes.go index 2382c575..2533036d 100644 --- a/backend/internal/api/routes/routes.go +++ b/backend/internal/api/routes/routes.go @@ -29,6 +29,29 @@ import ( _ "github.com/Wikid82/charon/backend/pkg/dnsprovider/custom" ) +type uptimeBootstrapService interface { + CleanupStaleFailureCounts() error + SyncMonitors() error + CheckAll() +} + +func runInitialUptimeBootstrap(enabled bool, uptimeService uptimeBootstrapService, logWarn func(error, string), logError func(error, string)) { + if !enabled { + return + } + + if err := uptimeService.CleanupStaleFailureCounts(); err != nil && logWarn != nil { + logWarn(err, "Failed to cleanup stale failure counts") + } + + if err := uptimeService.SyncMonitors(); err != nil && logError != nil { + logError(err, "Failed to sync monitors") + } + + // Run initial check immediately after sync to avoid the 90s blind window. + uptimeService.CheckAll() +} + // Register wires up API routes and performs automatic migrations. func Register(router *gin.Engine, db *gorm.DB, cfg config.Config) error { // Caddy Manager - created early so it can be used by settings handlers for config reload @@ -464,18 +487,12 @@ func RegisterWithDeps(router *gin.Engine, db *gorm.DB, cfg config.Config, caddyM enabled = s.Value == "true" } - if enabled { - // Clean up stale failure counts from historical bugs before first sync - if err := uptimeService.CleanupStaleFailureCounts(); err != nil { - logger.Log().WithError(err).Warn("Failed to cleanup stale failure counts") - } - - if err := uptimeService.SyncMonitors(); err != nil { - logger.Log().WithError(err).Error("Failed to sync monitors") - } - // Run initial check immediately after sync to avoid the 90s blind window - uptimeService.CheckAll() - } + runInitialUptimeBootstrap( + enabled, + uptimeService, + func(err error, msg string) { logger.Log().WithError(err).Warn(msg) }, + func(err error, msg string) { logger.Log().WithError(err).Error(msg) }, + ) ticker := time.NewTicker(1 * time.Minute) for range ticker.C { diff --git a/backend/internal/api/routes/routes_coverage_test.go b/backend/internal/api/routes/routes_coverage_test.go index e5e11d82..57939ce7 100644 --- a/backend/internal/api/routes/routes_coverage_test.go +++ b/backend/internal/api/routes/routes_coverage_test.go @@ -73,3 +73,55 @@ func TestRegister_LegacyMigrationErrorIsNonFatal(t *testing.T) { } require.True(t, hasHealth) } + +func TestRegister_UptimeFeatureFlagDefaultErrorIsNonFatal(t *testing.T) { + gin.SetMode(gin.TestMode) + router := gin.New() + + db, err := gorm.Open(sqlite.Open("file::memory:?cache=shared&_test_uptime_flag_warn"), &gorm.Config{ + Logger: logger.Default.LogMode(logger.Silent), + }) + require.NoError(t, err) + + const cbName = "routes:test_force_settings_query_error" + err = db.Callback().Query().Before("gorm:query").Register(cbName, func(tx *gorm.DB) { + if tx.Statement != nil && tx.Statement.Table == "settings" { + _ = tx.AddError(errors.New("forced settings query failure")) + } + }) + require.NoError(t, err) + t.Cleanup(func() { + _ = db.Callback().Query().Remove(cbName) + }) + + cfg := config.Config{JWTSecret: "test-secret"} + + err = Register(router, db, cfg) + require.NoError(t, err) +} + +func TestRegister_SecurityHeaderPresetInitErrorIsNonFatal(t *testing.T) { + gin.SetMode(gin.TestMode) + router := gin.New() + + db, err := gorm.Open(sqlite.Open("file::memory:?cache=shared&_test_sec_header_presets_warn"), &gorm.Config{ + Logger: logger.Default.LogMode(logger.Silent), + }) + require.NoError(t, err) + + const cbName = "routes:test_force_security_header_profile_query_error" + err = db.Callback().Query().Before("gorm:query").Register(cbName, func(tx *gorm.DB) { + if tx.Statement != nil && tx.Statement.Table == "security_header_profiles" { + _ = tx.AddError(errors.New("forced security_header_profiles query failure")) + } + }) + require.NoError(t, err) + t.Cleanup(func() { + _ = db.Callback().Query().Remove(cbName) + }) + + cfg := config.Config{JWTSecret: "test-secret"} + + err = Register(router, db, cfg) + require.NoError(t, err) +} diff --git a/backend/internal/api/routes/routes_uptime_bootstrap_test.go b/backend/internal/api/routes/routes_uptime_bootstrap_test.go new file mode 100644 index 00000000..ac03c221 --- /dev/null +++ b/backend/internal/api/routes/routes_uptime_bootstrap_test.go @@ -0,0 +1,107 @@ +package routes + +import ( + "errors" + "testing" + + "github.com/stretchr/testify/assert" +) + +type testUptimeBootstrapService struct { + cleanupErr error + syncErr error + + cleanupCalls int + syncCalls int + checkAllCalls int +} + +func (s *testUptimeBootstrapService) CleanupStaleFailureCounts() error { + s.cleanupCalls++ + return s.cleanupErr +} + +func (s *testUptimeBootstrapService) SyncMonitors() error { + s.syncCalls++ + return s.syncErr +} + +func (s *testUptimeBootstrapService) CheckAll() { + s.checkAllCalls++ +} + +func TestRunInitialUptimeBootstrap_Disabled_DoesNothing(t *testing.T) { + svc := &testUptimeBootstrapService{} + + warnLogs := 0 + errorLogs := 0 + runInitialUptimeBootstrap( + false, + svc, + func(err error, msg string) { warnLogs++ }, + func(err error, msg string) { errorLogs++ }, + ) + + assert.Equal(t, 0, svc.cleanupCalls) + assert.Equal(t, 0, svc.syncCalls) + assert.Equal(t, 0, svc.checkAllCalls) + assert.Equal(t, 0, warnLogs) + assert.Equal(t, 0, errorLogs) +} + +func TestRunInitialUptimeBootstrap_Enabled_HappyPath(t *testing.T) { + svc := &testUptimeBootstrapService{} + + warnLogs := 0 + errorLogs := 0 + runInitialUptimeBootstrap( + true, + svc, + func(err error, msg string) { warnLogs++ }, + func(err error, msg string) { errorLogs++ }, + ) + + assert.Equal(t, 1, svc.cleanupCalls) + assert.Equal(t, 1, svc.syncCalls) + assert.Equal(t, 1, svc.checkAllCalls) + assert.Equal(t, 0, warnLogs) + assert.Equal(t, 0, errorLogs) +} + +func TestRunInitialUptimeBootstrap_Enabled_CleanupError_StillProceeds(t *testing.T) { + svc := &testUptimeBootstrapService{cleanupErr: errors.New("cleanup failed")} + + warnLogs := 0 + errorLogs := 0 + runInitialUptimeBootstrap( + true, + svc, + func(err error, msg string) { warnLogs++ }, + func(err error, msg string) { errorLogs++ }, + ) + + assert.Equal(t, 1, svc.cleanupCalls) + assert.Equal(t, 1, svc.syncCalls) + assert.Equal(t, 1, svc.checkAllCalls) + assert.Equal(t, 1, warnLogs) + assert.Equal(t, 0, errorLogs) +} + +func TestRunInitialUptimeBootstrap_Enabled_SyncError_StillChecksAll(t *testing.T) { + svc := &testUptimeBootstrapService{syncErr: errors.New("sync failed")} + + warnLogs := 0 + errorLogs := 0 + runInitialUptimeBootstrap( + true, + svc, + func(err error, msg string) { warnLogs++ }, + func(err error, msg string) { errorLogs++ }, + ) + + assert.Equal(t, 1, svc.cleanupCalls) + assert.Equal(t, 1, svc.syncCalls) + assert.Equal(t, 1, svc.checkAllCalls) + assert.Equal(t, 0, warnLogs) + assert.Equal(t, 1, errorLogs) +} diff --git a/backend/internal/services/uptime_service_pr1_test.go b/backend/internal/services/uptime_service_pr1_test.go index dd3c97fd..162077ff 100644 --- a/backend/internal/services/uptime_service_pr1_test.go +++ b/backend/internal/services/uptime_service_pr1_test.go @@ -1,6 +1,7 @@ package services import ( + "errors" "fmt" "net/http" "net/http/httptest" @@ -246,6 +247,63 @@ func TestSyncAndCheckForHost_MissingSetting_StillCreates(t *testing.T) { assert.Greater(t, count, int64(0), "monitor should be created when setting is missing (default: enabled)") } +func TestSyncAndCheckForHost_UsesDomainWhenHostNameMissing(t *testing.T) { + db := setupPR1TestDB(t) + enableUptimeFeature(t, db) + svc := NewUptimeService(db, nil) + server := createAlwaysOKServer(t) + domain := hostPortFromServerURL(server.URL) + + host := createTestProxyHost(t, db, "", domain, "10.10.10.10") + + svc.SyncAndCheckForHost(host.ID) + + var monitor models.UptimeMonitor + require.NoError(t, db.Where("proxy_host_id = ?", host.ID).First(&monitor).Error) + assert.Equal(t, domain, monitor.Name) +} + +func TestSyncAndCheckForHost_CreateMonitorError_ReturnsWithoutPanic(t *testing.T) { + db := setupPR1TestDB(t) + enableUptimeFeature(t, db) + svc := NewUptimeService(db, nil) + server := createAlwaysOKServer(t) + domain := hostPortFromServerURL(server.URL) + + host := createTestProxyHost(t, db, "create-error-host", domain, "10.10.10.11") + + callbackName := "test:force_uptime_monitor_create_error" + require.NoError(t, db.Callback().Create().Before("gorm:create").Register(callbackName, func(tx *gorm.DB) { + if tx.Statement != nil && tx.Statement.Schema != nil && tx.Statement.Schema.Name == "UptimeMonitor" { + _ = tx.AddError(errors.New("forced uptime monitor create error")) + } + })) + t.Cleanup(func() { + _ = db.Callback().Create().Remove(callbackName) + }) + + assert.NotPanics(t, func() { + svc.SyncAndCheckForHost(host.ID) + }) + + var count int64 + db.Model(&models.UptimeMonitor{}).Where("proxy_host_id = ?", host.ID).Count(&count) + assert.Equal(t, int64(0), count) +} + +func TestSyncAndCheckForHost_QueryMonitorError_ReturnsWithoutPanic(t *testing.T) { + db := setupPR1TestDB(t) + enableUptimeFeature(t, db) + svc := NewUptimeService(db, nil) + host := createTestProxyHost(t, db, "query-error-host", "query-error.example.com", "10.10.10.12") + + require.NoError(t, db.Migrator().DropTable(&models.UptimeMonitor{})) + + assert.NotPanics(t, func() { + svc.SyncAndCheckForHost(host.ID) + }) +} + // --- Fix 4: CleanupStaleFailureCounts --- func TestCleanupStaleFailureCounts_ResetsStuckMonitors(t *testing.T) { @@ -360,6 +418,19 @@ func TestCleanupStaleFailureCounts_DoesNotResetDownHosts(t *testing.T) { assert.Equal(t, "down", h.Status, "cleanup must not reset host status") } +func TestCleanupStaleFailureCounts_ReturnsErrorWhenDatabaseUnavailable(t *testing.T) { + db := setupPR1TestDB(t) + svc := NewUptimeService(db, nil) + + sqlDB, err := db.DB() + require.NoError(t, err) + require.NoError(t, sqlDB.Close()) + + err = svc.CleanupStaleFailureCounts() + require.Error(t, err) + assert.Contains(t, err.Error(), "cleanup stale failure counts") +} + // setupPR1ConcurrentDB creates a file-based SQLite database with WAL mode and // busy_timeout to handle concurrent writes without "database table is locked". func setupPR1ConcurrentDB(t *testing.T) *gorm.DB { diff --git a/frontend/src/pages/__tests__/Uptime.test.tsx b/frontend/src/pages/__tests__/Uptime.test.tsx index 53776e7b..96b0e93d 100644 --- a/frontend/src/pages/__tests__/Uptime.test.tsx +++ b/frontend/src/pages/__tests__/Uptime.test.tsx @@ -139,6 +139,23 @@ describe('Uptime page', () => { expect(screen.getByText('Loading monitors...')).toBeInTheDocument() }) + it('falls back to DOWN status when monitor status is unknown', async () => { + const { getMonitors, getMonitorHistory } = await import('../../api/uptime') + const monitor = { + id: 'm-unknown-status', name: 'UnknownStatusMonitor', url: 'http://example.com', type: 'http', interval: 60, enabled: true, + status: 'mystery', last_check: new Date().toISOString(), latency: 10, max_retries: 3, + } + vi.mocked(getMonitors).mockResolvedValue([monitor]) + vi.mocked(getMonitorHistory).mockResolvedValue([]) + + renderWithQueryClient() + await waitFor(() => expect(screen.getByText('UnknownStatusMonitor')).toBeInTheDocument()) + + const badge = screen.getByTestId('status-badge') + expect(badge).toHaveAttribute('data-status', 'down') + expect(badge).toHaveTextContent('DOWN') + }) + it('renders empty state when no monitors exist', async () => { const { getMonitors } = await import('../../api/uptime') vi.mocked(getMonitors).mockResolvedValue([])