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

Fix vsc backup plugin progress method to not fail on temporary errors #226

Draft
wants to merge 1 commit into
base: main
Choose a base branch
from
+3 −2
Draft

Fix vsc backup plugin progress method to not fail on temporary errors #226

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
5 changes: 3 additions & 2 deletions internal/backup/volumesnapshotcontent_action.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -153,10 +153,11 @@ func (p *VolumeSnapshotContentBackupItemAction) Progress(operationID string, bac
if boolptr.IsSetToTrue(vsc.Status.ReadyToUse) {
progress.Completed = true
} else if vsc.Status.Error != nil {
progress.Completed = true
errorMessage := ""
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If we don't mark the progress.Completed as true when there is an error in the VSC and VS status, one possible side-effect is that if the error cannot be resolved by retrying, the action will not be completed until the timeout. The default async operation timeout is 4 hours.

Copy link
Contributor

@blackpiglet blackpiglet Jan 30, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we add a mechanism to count down the failure limitation?
For example, we can set a max try time for the VSC on error, and put that information in the VSC annotation, then update the error tried number accordingly. Until the limitation is hit, fail the action.

The default async operation sync frequency is 10s. We can introduce a new parameter alongside --item-operation-sync-frequency to set the retry limitation.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I am not sure how to track for number of errors, since last error might stay same and not be updated.
Might be better to somehow enforce 10min limit and then mark it as failed.

Copy link
Contributor

@blackpiglet blackpiglet Jan 31, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks. A timer is better. Propose the name as item-operation-error-timeout.

The timer should be compared with the VolumeSnapshotContent.Status.Error.Time.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@blackpiglet @anshulahuja98 the new parameter --item-operation-error-timeout sounds good, will work on a PR for velero controller and then update this PR accordingly.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this parameter however might lead to more concerns
where each plugin might want to have a different time limit, whereas this is a global setting

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It is better if somehow we give this error exit control to the plugin to be smart enough to exit by it's own logic than a global setting

Copy link
Contributor

@blackpiglet blackpiglet Feb 1, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just realized I didn't think it through.
It's not possible to have a global timer work in all of the plugins.
item-operation-error-timeout works globally, because it's triggered in the Velero server.

If we don't modify the BIA and RIA interface, we cannot pass the timer to the Progress method.

item-operation-error-timeout can only work in the backup_operations_controllers.go. Looks like that is not what we want to do.

Adding a default timer in the plugin code is the simplest way, but we cannot modify the timer.
Another possible way is adding the Init method just like the ObjectStore and the VolumeSnapshotter plugins.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@blackpiglet @anshulahuja98 If we add item-operation-error-timeout to backup spec then we can use it in BIA progress method, right ?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@shubham-pampattiwar
That should work.

if vsc.Status.Error.Message != nil {
progress.Err = *vsc.Status.Error.Message
errorMessage = *vsc.Status.Error.Message
}
p.Log.Warnf("VolumeSnapshotContent has a temporary error %s. SnapshotContent controller will retry later.", errorMessage)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
p.Log.Warnf("VolumeSnapshotContent has a temporary error %s. SnapshotContent controller will retry later.", errorMessage)
p.Log.Warnf("VolumeSnapshotContent has a temporary error: %s. SnapshotContent controller will retry later.", errorMessage)

}

return progress, nil
Expand Down
Loading