-
Notifications
You must be signed in to change notification settings - Fork 382
Allow retries for instances with Failed condition after spec changes #1751
Allow retries for instances with Failed condition after spec changes #1751
Conversation
@@ -110,7 +110,7 @@ func withConfigGetFreshApiserverAndClient( | |||
|
|||
if err := server.RunServer(options, stopCh); err != nil { | |||
close(serverFailed) | |||
t.Fatalf("Error in bringing up the server: %v", err) | |||
t.Logf("Error in bringing up the server: %v", err) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
see golang/go#15976
The integration test was failing with data race
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We should also remove the Failed
condition when we're starting the new operation.
This needs to wait on #1748, right? Otherwise, if the user tries to update an instance that failed to provision, then the controller will attempt to send an update request to the broker, instead of a provision request. |
@kibbles-n-bytes yes, you're right, I do this in #1748 |
The only issue with allowing retries for instances with See #1771 (comment) |
LGTM |
…ubernetes-retired#1751) * Allow retries for instances with Failed condition after spec changes * Remove Fatalf from server startup goroutine to fix data race
Implementation of item number
2.
from the checklist #1715 (comment)After this change we should be able to retry after receiving
400 Bad Request
from broker as soon as we change the spec.