Repository navigation
fix(web-ui): quarantine is not a fault (#1076) + unblock Windows E2E (#1034) - #1077
Merged
Dumbris merged 1 commit intoAug 28, 2026
Merged
Conversation
…1034) Two independent fixes, both clearing red state ahead of the v0.62.0 cut. #1076 — Server Detail rendered the red ErrorPanel AND the calm Security Quarantine banner off one payload. mcpproxy still attempts a connection to a quarantined server so the scanner can export its tool definitions; that attempt fails and leaves an error-severity diagnostic behind — usually the unclassified one — while internal/health/calculator.go short-circuits the server to healthy/quarantined. Both facts are true in one response, and the UI read only the first, telling the user to file a bug report about a server that is simply waiting for them to approve it. Quarantine now suppresses every fault alert on the page, following the SignInPanel precedent one block up (MCP-1821): an intentional admin state is not a fault. The guard covers the generic `last_error` alert as well as the diagnostic panel — without that, suppressing one just falls through to the other and prints the same fault in a different red box. This is the same rule #1075 applied to the tray badge, on the surface that PR deliberately did not touch; `admin_state` stays the cross-surface contract. #1034 — the E2E Tests / windows-latest job has been red on main since 2026-08-13. Root cause is a timeout-budget mismatch between two workflows running the same sweep: e2e-tests.yml used `-timeout 4m` where unit-tests.yml uses 10m. internal/server took 240s on the Windows runner and died at `panic: test timed out after 4m0s` with a single test only 2s in — cumulative slowness (Bleve index creation on the Windows filesystem), never a hang, so raising the budget to match is the whole fix. Verified: 959/959 frontend unit tests, vue-tsc clean, and the Playwright web-UI sweep run four ways (with and without the fixture upstream, on this branch and on a rebuilt origin/main baseline) — identical results each pairing, so the sweep's remaining failures are pre-existing on main and untouched by this change.
Deploying mcpproxy-docs with
|
| Latest commit: |
7606769
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://8122e63d.mcpproxy-docs.pages.dev |
| Branch Preview URL: | https://fix-1076-quarantine-error-pa.mcpproxy-docs.pages.dev |
|
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
Contributor
📦 Build ArtifactsWorkflow Run: View Run Available Artifacts
How to DownloadOption 1: GitHub Web UI (easiest)
Option 2: GitHub CLI gh run download 33180183733 --repo smart-mcp-proxy/mcpproxy-go
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Two independent fixes, both clearing red state ahead of the v0.62.0 cut.
Closes #1076
Closes #1034
#1076 — quarantined server told to file a bug report
Server Detail rendered the red
ErrorPaneland the calm Security Quarantine banner off one payload that sayshealthy/quarantined.mcpproxy still attempts a connection to a quarantined server so the scanner can export its tool definitions. That attempt fails and leaves an error-severity diagnostic behind — usually
MCPX_UNKNOWN_UNCLASSIFIED, whose copy asks the user to file a bug report — whileinternal/health/calculator.goshort-circuits the server tolevel: healthy,admin_state: quarantined,action: approve. Both facts are true in one response; the UI read only the first.Quarantine now suppresses every fault alert on the page, following the
SignInPanelprecedent one block up (MCP-1821): an intentional admin state is not a fault, and the quarantine banner already states the situation and the action.The guard deliberately covers both red surfaces — the diagnostic panel and the generic
v-else-if="server.last_error"alert below it. The issue flagged this: suppressing only the first falls straight through to the second and prints the same fault in a different red box.This is the rule #1075 applied to the tray badge (
ServerStatus.isBadgeExempt), now on the surface that PR deliberately did not touch.admin_statestays the cross-surface contract.Scope is unchanged elsewhere: a non-quarantined server with the same diagnostic still gets the full
ErrorPanel, andSignInPanelkeeps its precedence for a quarantined server that needs sign-in.#1034 — E2E Tests / windows-latest, red on main since 2026-08-13
A timeout-budget mismatch between two workflows running the same sweep:
e2e-tests.ymlused-timeout 4mwhereunit-tests.ymluses10m.internal/servertook 240s on the Windows runner and died atpanic: test timed out after 4m0s.Nothing hangs — the panic's
running tests:line named a single test only 2s in (TestProfile_UnauthFilteredByProfile), so this is cumulative slowness (Bleve index creation on the Windows filesystem), and raising the budget to match the sibling workflow is the whole fix. A comment on the step records that, so the two budgets are kept in step.The nil-deref panics visible in the same job log are unrelated and not a production bug:
agent_token_gating_test.godrives every mutating route throughbaseController, whoseEditRegistrySourceRefreturns(nil, nil, nil); chi's Recoverer catches it and the auth-gate assertion still passes.Verification
frontend/tests/unit/server-detail-quarantine-error-panel.spec.ts— 5 cases, written first and watched fail for the right reasons (ErrorPanel present on a quarantined server, with and withoutlast_error). Two of the five passed from the start as deliberate regression guards on the narrowness of the fix.vue-tsc --noEmitclean.scripts/run-web-smoke.sh, required forfrontend/src/changes) run four ways: with and withoutMCPPROXY_FIXTURE_PATH, on this branch and on a rebuiltorigin/mainbaseline. Each pairing produced identical results, so the sweep's remaining failures are pre-existing on main and untouched here. Theweb-ui-sweep.spec.tssuite — including server detail — passes.The pre-existing sweep failures are worth their own issue and are not addressed here: a 4.23:1 contrast on the dashboard empty-state paragraph (
opacity-60, below AA), and three data-dependentvisual-a11y-sweepchecks that only fail once a fixture upstream is registered.