Skip to content

test/e2e/upgrade: Relax 'too long' soft timeout for rollback jobs #25977

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
openshift-merge-robot merged 1 commit into from
Mar 19, 2021
Merged

test/e2e/upgrade: Relax 'too long' soft timeout for rollback jobs #25977

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
12 changes: 8 additions & 4 deletions test/e2e/upgrade/upgrade.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -277,9 +277,11 @@ func clusterUpgrade(f *framework.Framework, c configv1client.Interface, dc dynam
case upgradeAbortAtRandom:
abortAt = int(rand.Int31n(100) + 1)
maximumDuration *= 2
durationToSoftFailure *= 2
framework.Logf("Upgrade will be aborted and the cluster will roll back to the current version after %d%% of operators have upgraded (picked randomly)", abortAt)
default:
maximumDuration *= 2
durationToSoftFailure *= 2
framework.Logf("Upgrade will be aborted and the cluster will roll back to the current version after %d%% of operators have upgraded", upgradeAbortAt)
}

Expand Down Expand Up @@ -349,6 +351,7 @@ func clusterUpgrade(f *framework.Framework, c configv1client.Interface, dc dynam
func() error {
framework.Logf("Cluster version operator acknowledged upgrade request")
aborted := false
action := "upgrade"
var lastMessage string
upgradeStarted := time.Now()

Expand Down Expand Up @@ -383,26 +386,27 @@ func clusterUpgrade(f *framework.Framework, c configv1client.Interface, dc dynam
}
recordClusterEvent(kubeClient, uid, "Upgrade", "UpgradeRollback", fmt.Sprintf("version/%s image/%s", original.Status.Desired.Version, original.Status.Desired.Version), false)
aborted = true
action = "aborted upgrade"
return false, nil
}

return monitor.Reached(cv, desired)

}); err != nil {
if lastMessage != "" {
return fmt.Errorf("Cluster did not complete upgrade: %v: %s", err, lastMessage)
return fmt.Errorf("Cluster did not complete %s: %v: %s", action, err, lastMessage)
}
return fmt.Errorf("Cluster did not complete upgrade: %v", err)
return fmt.Errorf("Cluster did not complete %s: %v", action, err)
}

framework.Logf("Completed upgrade to %s", versionString(desired))
framework.Logf("Completed %s to %s", action, versionString(desired))
recordClusterEvent(kubeClient, uid, "Upgrade", "UpgradeVersion", fmt.Sprintf("version/%s image/%s", updated.Status.Desired.Version, updated.Status.Desired.Version), false)

// record whether the cluster was fast or slow upgrading. Don't fail the test, we still want signal on the actual tests themselves.
upgradeEnded := time.Now()
upgradeDuration := upgradeEnded.Sub(upgradeStarted)
if upgradeDuration > durationToSoftFailure {
disruption.RecordJUnitResult(f, "[sig-cluster-lifecycle] cluster upgrade should be fast", upgradeDuration, fmt.Sprintf("Upgrade took too long: %v", upgradeDuration.Minutes()))
disruption.RecordJUnitResult(f, "[sig-cluster-lifecycle] cluster upgrade should be fast", upgradeDuration, fmt.Sprintf("%s to %s took too long: %v", action, versionString(desired), upgradeDuration.Minutes()))
} else {
disruption.RecordJUnitResult(f, "[sig-cluster-lifecycle] cluster upgrade should be fast", upgradeDuration, "")
}
Expand Down