Re: [kubernetes/kubernetes] TestCascadingDeletion flaky [timed out waiting for caches to sync] (#56121)

1 view

Skip to first unread message

Jordan Liggitt

unread,

Nov 28, 2017, 11:40:58 PM11/28/17

to kubernetes/kubernetes, k8s-mirror-api-machinery-test-failures, Team mention

@kubernetes/sig-api-machinery-test-failures

—
You are receiving this because you are on a team that was mentioned.
Reply to this email directly, view it on GitHub, or mute the thread.

Dan Mace

unread,

Nov 29, 2017, 9:19:12 AM11/29/17

to kubernetes/kubernetes, k8s-mirror-api-machinery-test-failures, Team mention

Pretty sure this is a long standing flake, but I'm not yet sure yet which issue to consider it a duplicate of. Some previous reports are #28713, #27460, and #25890. The timeout for the failing condition was at some point reduced from 120 to 60 seconds in 4d23506 without any explanation in the commit message.

I'm guessing @caesarxuchao has some more of the history of this one to share with us. A search of prior issues reveals multiple attempted fixes.

P.S., the title and description of this issue seem inaccurate. The failure related to TestCascadingDeletion is at

garbage_collector_test.go:366: timed out waiting for the condition