node_e2e: refactor RunTogether function #124668
Conversation
|
Skipping CI for Draft Pull Request. |
k8s-ci-robot
added
release-note-none
|
/sig node |
k8s-ci-robot
added
area/test
sig/node
|
/test all |
|
/test pull-kubernetes-node-swap-fedora |
bart0sh
force-pushed
the
PR143-e2e-node-fix-containers-lifecycle
branch
from
db583df
to
6ecf0da
Compare
|
/test pull-kubernetes-node-swap-fedora |
|
/test all |
|
So far so good: https://prow.k8s.io/view/gs/kubernetes-jenkins/pr-logs/directory/pull-kubernetes-node-swap-fedora/1785987645336195072 I'm going to trigger this job again and again to see if it flakes or not. |
|
/triage accepted |
k8s-ci-robot
added
triage/accepted
|
/assign @SergeyKanzhelev |
|
/test pull-kubernetes-node-swap-fedora |
2 similar comments
|
/test pull-kubernetes-node-swap-fedora |
|
/test pull-kubernetes-node-swap-fedora |
|
/test pull-kubernetes-node-swap-fedora-serial |
| if lhsStart == -1 { | ||
| return fmt.Errorf("couldn't find that %s ever started, got\n%v", lhs, o) | ||
| } | ||
|
|
||
| rhsStart := o.findIndex(rhs, "Started", lhsStart+1) |
There was a problem hiding this comment.
This will exclude the case that rhs starts before the lhs but they run together.
Could you explain what is the reason for changing this?
There was a problem hiding this comment.
this is more reproducible IMHO and if rhs starts first, just swap rhs and lhs (as it's done in the other file)
There was a problem hiding this comment.
I am a little bit worried about the ordering part as we cannot use this in cases that we cannot guarantee the ordering (e.g. regular containers run together part?).
How about adding another test function like RunTogetherWithOrdering instead?
There was a problem hiding this comment.
well today we don't use that with regular containers... so we could add another one RunTogetherWithoutOrdering?
There was a problem hiding this comment.
Yes, i am ok with that.
Anyway, I am just curious... could you explain why adding the ordering part deflakes the tests?
There was a problem hiding this comment.
it's to avoid weird situations when containers are restarting in a loop
There was a problem hiding this comment.
Sure, we expect that the first container is started before the second one as they must be started sequentially.
|
/test pull-kubernetes-node-kubelet-serial-containerd-sidecar-containers |
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: bart0sh, matthyx, SergeyKanzhelev The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
k8s-ci-robot
added
the
approved
|
/test pull-kubernetes-node-swap-fedora |
|
@bart0sh: The following test failed, say
Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here. |
|
/skip |
|
/skip pull-kubernetes-node-swap-fedora-serial |
What type of PR is this?
/kind bug
/kind cleanup
/kind flake
What this PR does / why we need it:
This is a follow up PR for #124645 aiming to fix flaky container lifecycle test cases.
Which issue(s) this PR fixes:
Special notes for your reviewer:
Does this PR introduce a user-facing change?