Skip to content

feat: Introduce count thresholds for unrecoverable pod events #795

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Mar 17, 2022
Merged

feat: Introduce count thresholds for unrecoverable pod events #795

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
28 changes: 19 additions & 9 deletions pkg/provision/workspace/deployment.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,12 +51,16 @@ var containerFailureStateReasons = []string{
"RunContainerError",
}

var unrecoverablePodEventReasons = []string{
"FailedPostStartHook",
"FailedMount",
"FailedScheduling",
"FailedCreate",
"ReplicaSetCreateError",
// unrecoverablePodEventReasons contains Kubernetes events that should fail workspace startup
// if they occur related to a workspace pod. Events are stored as a map with event names as keys
// and values representing the threshold of how many times we can see an event before it is considered
// unrecoverable.
var unrecoverablePodEventReasons = map[string]int32{
"FailedPostStartHook": 1,
"FailedMount": 3,
"FailedScheduling": 1,
"FailedCreate": 1,
"ReplicaSetCreateError": 1,
}

var unrecoverableDeploymentConditionReasons = []string{
Expand Down Expand Up @@ -473,9 +477,15 @@ func checkPodEvents(pod *corev1.Pod, workspaceID string, clusterAPI sync.Cluster
continue
}

for _, fatalEv := range unrecoverablePodEventReasons {
if ev.Reason == fatalEv && !checkIfUnrecoverableEventIgnored(ev.Reason) {
return fmt.Sprintf("Detected unrecoverable event %s: %s", ev.Reason, ev.Message), nil
if maxCount, isUnrecoverableEvent := unrecoverablePodEventReasons[ev.Reason]; isUnrecoverableEvent {
if !checkIfUnrecoverableEventIgnored(ev.Reason) && ev.Count >= maxCount {
var msg string
if ev.Count > 1 {
msg = fmt.Sprintf("Detected unrecoverable event %s %d times: %s", ev.Reason, ev.Count, ev.Message)
} else {
msg = fmt.Sprintf("Detected unrecoverable event %s: %s", ev.Reason, ev.Message)
}
return msg, nil
}
}
}
Expand Down