Fix clean_jobs missing --namespace flag causing BackoffLimitExceeded

[circleci-app]

Problem

deploy_to_doks_release (job 30881) failed at the "Install Codacy" step with:

Error: UPGRADE FAILED: release codacy-release failed, and has been rolled back due to atomic being set: post-upgrade hooks failed: job failed: BackoffLimitExceeded

The previous fix (#836) added a clean_jobs Makefile target to delete stale hook jobs before the helm upgrade. However, that target ran:

kubectl delete jobs -l app.kubernetes.io/instance=codacy-release --ignore-not-found=true

…without a --namespace flag, so kubectl searched the default namespace instead of codacy-release. The command returned "No resources found" and the stale hook jobs (DB migrations, etc.) in codacy-release were never deleted. On the next helm upgrade attempt, helm found the existing BackoffLimitExceeded job and immediately failed again.

Fix

Add --namespace ${NAMESPACE} to the kubectl delete jobs command in the clean_jobs target so it targets the correct namespace for every environment (dev, sandbox, release).

clean_jobs:
    -kubectl delete jobs -l app.kubernetes.io/instance=${RELEASE_NAME} --namespace ${NAMESPACE} --ignore-not-found=true

Test

Re-running deploy_to_doks_release should now successfully delete stale hook jobs in the codacy-release namespace before the helm upgrade proceeds.

https://app.circleci.com/agents/gh/codacy/chat/53169a49-798b-45fd-a1cc-9721760e5036

[codacy-production]

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

AI Reviewer: first review requested successfully. AI can make mistakes. Always validate suggestions.

_{TIP This summary will be updated as you push new changes.}

[codacy-production]

Pull Request Overview

The PR addresses a deployment blocker by adding the missing --namespace flag to the clean_jobs Kubernetes deletion command, ensuring stale jobs are cleared from the correct environment. Codacy analysis indicates that the changes are up to standards with no new quality issues or complexity regressions.

While the logic appears sound, there is a gap in automated verification. A required test scenario to confirm the correct population of the namespace and label selector variables in the kubectl command is currently missing. Implementing this verification is recommended to prevent future regressions in environment-specific cleanup tasks.

Test suggestions

• Verify that clean_jobs executes kubectl delete with the correct namespace and label selector variables

Prompt proposal for missing tests

Consider implementing these tests if applicable:
1. Verify that clean_jobs executes kubectl delete with the correct namespace and label selector variables

_{TIP Improve review quality by adding custom instructions
TIP How was this review? Give us feedback}