cleanup test namespace #42
Conversation
![codereviewbot-ai[bot]](https://avatars.githubusercontent.com/in/309638?s=60&v=4){kind=small}
| set +e | ||
|
|
||
| ns_cleanup() { | ||
| echo "Performing namespace ${NAMESPACE} cleanup..." | ||
| ${KUBECTL} --kubeconfig=${KUBECONFIG} --cluster=${CLUSTER} --user=${USER} delete namespace --wait=false ${NAMESPACE} | ||
| } |
There was a problem hiding this comment.
The set +e command disables the immediate exit on error, which can lead to the script continuing execution even if critical commands fail. This can cause unexpected behavior and make debugging difficult. Consider handling errors explicitly where needed and re-enabling set -e after the critical section.
| @@ -71,6 +72,8 @@ else | |||
| exit 1 | |||
| fi | |||
| done | |||
There was a problem hiding this comment.
The while loop that generates a random namespace does not have a termination condition if the namespace creation keeps failing. This can lead to an infinite loop. Consider adding a maximum retry limit to avoid potential infinite loops.
| go func() { | ||
| err := stern.Run(ctx, *namespace, clientset, allowErrors, disablePodLogs) |
There was a problem hiding this comment.
The goroutine starting at line 374 calls stern.Run but does not handle any potential errors returned by this function. If stern.Run fails, the error will be lost, making debugging difficult.
Recommendation: Handle the error returned by stern.Run within the goroutine, possibly by logging it or sending it to a channel for further processing.
kubectl failure in integration test setup script causes the script to exit prematurely so it_sidecar does not have a chance to perform the cleanup.
This PR adds a trap that would be executed on exit of the script whether it normal exit or by error. This also make the cleanup part of the it_sidecar obsolete.