github/kubernetes
4
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix e2e test flakes.

Browse Source 
Makes number of failures per pod fixed at 1, for the RestartOnFailure
case, which prevents Kubelet restart backoff, which causes test timeout.
For RestartNever tests, it keeps using the random success/failure.

Fixes #15389.
Renables previously flaky e2e.
This commit is contained in:
Eric Tune
2015-10-24 22:51:02 -07:00
parent a094a6e3de
commit 40a33dcc7e
2 changed files with 34 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
hack/jenkins/e2e.sh
View File

@@ -135,7 +135,6 @@ DISRUPTIVE_TESTS=(
GCE_FLAKY_TESTS=(
"DaemonRestart\sController\sManager"
"Daemon\sset\sshould\srun\sand\sstop\scomplex\sdaemon"
"Jobs\sare\slocally\srestarted"
"Resource\susage\sof\ssystem\scontainers"
"allows\sscheduling\sof\spods\son\sa\sminion\safter\sit\srejoins\sthe\scluster" # file: resize_nodes.go, issue: #13258
"pod\sw/two\sRW\sPDs\sboth\smounted\sto\sone\scontainer,\swrite\sto\sPD" # file: pd.go, issue: #15382