Repository navigation
SF stability check: wait while a recreated app is not yet deployed on its node - #25
Conversation
…in the stability check Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
Codex Review Summary: ✅ CompletedThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: f89d84604a
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| if ($notDeployedOn.Count -gt 0) { | ||
| $state = 'NotDeployed' |
There was a problem hiding this comment.
Preserve Error health when a node has not deployed the application
If application health is Error while a node returns FabricElementNotFoundException, this assignment replaces the actual health with NotDeployed. If that condition persists until the readiness timeout, Wait-SfApplicationStable returns NeverOk with Health = 'NotDeployed', so Test-SfRollbackWarranted returns false and a configured rollback is skipped despite the application's Error health. Preserve Error when marking incomplete deployment, or track deployment readiness separately so it does not suppress the rollback decision.
Useful? React with 👍 / 👎.
Problema: tras recrear la app, Get-ServiceFabricDeployedCodePackage responde 'Application not found' en un nodo que ya tiene réplica y el chequeo abortaba (falso positivo; Payroll run 43927657408, app Ready/Ok).
Cambio: ese caso devuelve un snapshot NotDeployed y el chequeo sigue esperando hasta ReadyTimeoutSeconds; cualquier otro error sigue abortando y un reinicio o Health no Ok en la ventana sigue fallando.
Probado: Test-ServiceFabricStability.ps1 local (casos nuevos incluidos). Runners: el workflow no cambia, sigue self-hosted.
🤖 Generated with Claude Code
CI: repo público sin runner propio que lo acepte (decision-fixtures queda en cola). Igual que en #24, se corrió localmente
pwsh -File .github/actions/verify-servicefabric-stable/tests/Test-ServiceFabricStability.ps1: todos los casos PASS, incluidos los 4 nuevos.