Skip to content
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

Jump to bottom

Fix operator panic when checking for pod crashloop status #115

Merged
merged 3 commits into from
Oct 24, 2023
Merged

Fix operator panic when checking for pod crashloop status #115

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c56d4ad
check index out of bounds to prevent panic
SatyaKuppam Oct 21, 2023
627a8a7
simplify ordered ready crash loop checks
SatyaKuppam Oct 23, 2023
00ba60d
add else statement
SatyaKuppam Oct 23, 2023
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
35 changes: 22 additions & 13 deletions controllers/druid/handler.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -657,22 +657,31 @@ func checkCrashStatus(ctx context.Context, sdk client.Client, drd *v1alpha1.Drui
return err
}

// the below condition evalutes if a pod is in
// 1. failed state 2. unknown state
// OR condtion.status is false which evalutes if neither of these conditions are met
// 1. ContainersReady 2. PodInitialized 3. PodReady 4. PodScheduled
for _, p := range podList {
if p.(*v1.Pod).Status.ContainerStatuses[0].RestartCount > 1 {
if p.(*v1.Pod).Status.Phase == v1.PodFailed || p.(*v1.Pod).Status.Phase == v1.PodUnknown {
err := writers.Delete(ctx, sdk, drd, p, emitEvents, &client.DeleteOptions{})
if err != nil {
return err
}
msg := fmt.Sprintf("Deleted pod [%s] in namespace [%s], since it was in [%s] state.", p.GetName(), p.GetNamespace(), p.(*v1.Pod).Status.Phase)
logger.Info(msg, "Object", stringifyForLogging(p, drd), "name", drd.Name, "namespace", drd.Namespace)
} else {
for _, condition := range p.(*v1.Pod).Status.Conditions {
// condition.type Ready means the pod is able to service requests
if condition.Type == v1.ContainersReady {
// the below condition evalutes if a pod is in
// 1. pending state 2. failed state 3. unknown state
// OR condtion.status is false which evalutes if neither of these conditions are met
// 1. ContainersReady 2. PodInitialized 3. PodReady 4. PodScheduled
if p.(*v1.Pod).Status.Phase != v1.PodRunning || condition.Status == v1.ConditionFalse {
err := writers.Delete(ctx, sdk, drd, p, emitEvents, &client.DeleteOptions{})
if err != nil {
return err
} else {
msg := fmt.Sprintf("Deleted pod [%s] in namespace [%s], since it was in crashloopback state.", p.GetName(), p.GetNamespace())
logger.Info(msg, "Object", stringifyForLogging(p, drd), "name", drd.Name, "namespace", drd.Namespace)
if condition.Status == v1.ConditionFalse {
for _, containerStatus := range p.(*v1.Pod).Status.ContainerStatuses {
if containerStatus.RestartCount > 1 {
err := writers.Delete(ctx, sdk, drd, p, emitEvents, &client.DeleteOptions{})
if err != nil {
return err
}
msg := fmt.Sprintf("Deleted pod [%s] in namespace [%s], since the container [%s] was crashlooping.", p.GetName(), p.GetNamespace(), containerStatus.Name)
logger.Info(msg, "Object", stringifyForLogging(p, drd), "name", drd.Name, "namespace", drd.Namespace)
}
}
}
}
Expand Down
Loading