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.

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

Incremental improvements to kubelet e2e tests #24426

Merged
merged 1 commit into from
Apr 20, 2016

Conversation

pwittrock
Copy link
Member

  • Add keep-alive to ssh connection
  • Don't try to stop services on image-based runs
  • Increase jenkins ci timeout to 90 minutes to accomadate unpredictable go build times
  • Remove spammy log statement

- Add keep-alive to ssh connection
- Don't try to stop services on image-based runs
- Increase jenkins ci timeout to 90 minutes to accomadate unpredictable go build times
- Remove spammy log statement
@k8s-github-robot k8s-github-robot added size/S Denotes a PR that changes 10-29 lines, ignoring generated files. release-note-label-needed labels Apr 18, 2016
@@ -151,7 +150,6 @@ func (config *PrivilegedPodTestConfig) dialFromContainer(containerIP string, con
var output map[string]string
err = json.Unmarshal([]byte(stdout), &output)
Expect(err).NotTo(HaveOccurred(), fmt.Sprintf("Could not unmarshal curl response: %s", stdout))
glog.Infof("Deserialized output is %v", output)
Copy link
Contributor

Choose a reason for hiding this comment

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

This output was just not useful at all?

Copy link
Member Author

Choose a reason for hiding this comment

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

cc @yujuhong

The tests that call this log the map if it does not match expectations. I don't think we want to output this when the tests pass since it clutters up the build log.

Copy link
Contributor

Choose a reason for hiding this comment

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

I copied this from the original test. The output seems pretty meaningful. E.g., "privileged_test.go:154] Deserialized output is map[output:RTNETLINK answers: Operation not permitted"

Do we have any general rules for logging in a node e2e suite? A cluster e2e test logs quite a lot at the info level, but I've just noticed that it's very different in the node e2e suite.

Copy link
Member Author

Choose a reason for hiding this comment

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

I expect that message is ignored unless the test fails, in which case it should be printed out by the assertion. That output is actually expected, but can make debugging unrelated failures hard because it looks like an error message even though it is printed at info level.

My opinion would be that anything besides the test results themselves are just debug information that we should only assume will be seen if there is a test failure. To the extent possible, debug information should only be printed if it is related to a failed tests.

Copy link
Contributor

Choose a reason for hiding this comment

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

I agree in general that we should keep the output to the minimum, but fwiw, comparing the logs of a passed test against those of a failed test is helpful at times. I guess we won't need that if every test is well-written :)

@k8s-bot
Copy link

k8s-bot commented Apr 18, 2016

GCE e2e build/test passed for commit 90d2f9a.

@eparis eparis added lgtm "Looks good to me", indicates that a PR is ready to be merged. release-note Denotes a PR that will be considered when it comes time to generate release notes. and removed release-note-label-needed labels Apr 19, 2016
@pwittrock pwittrock added priority/important-soon Must be staffed and worked on either currently, or very soon, ideally in time for the next release. kind/flake Categorizes issue or PR as related to a flaky test. labels Apr 19, 2016
@pwittrock
Copy link
Member Author

Added p1 label since this is causing flaky failures and should recieve higher priority

@eparis
Copy link
Contributor

eparis commented Apr 19, 2016

is the flake fix just the timeout extension?

@pwittrock
Copy link
Member Author

No, there are a couple other flakes I am hoping this will help tease out.

#24423 CoreOs is kill the ssh connection mid test - I am seeing if the keepalive may help with this
#24422 Ginkgo tests exit -1 after suite passes - I prevented stoping the processes to see if this is causing a failure

@k8s-github-robot
Copy link

@k8s-bot test this [submit-queue is verifying that this PR is safe to merge]

@k8s-github-robot
Copy link

@k8s-bot test this issue: #IGNORE

Tests have been pending for 24 hours

@k8s-bot
Copy link

k8s-bot commented Apr 20, 2016

GCE e2e build/test passed for commit 90d2f9a.

@k8s-github-robot
Copy link

Automatic merge from submit-queue

@k8s-github-robot k8s-github-robot merged commit 86544c2 into kubernetes:master Apr 20, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
kind/flake Categorizes issue or PR as related to a flaky test. lgtm "Looks good to me", indicates that a PR is ready to be merged. priority/important-soon Must be staffed and worked on either currently, or very soon, ideally in time for the next release. release-note Denotes a PR that will be considered when it comes time to generate release notes. size/S Denotes a PR that changes 10-29 lines, ignoring generated files.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

7 participants