[release-1.18] managed_ns: deflake tests #3847
Conversation
turns out, the PID 42 (as idiomatic a test variable it is) is not the best PID to use. NamespacePaths() returns /proc entries if the PID is running, so sometimes there was a PID 42 on the system, and the tests returned valid paths, causing the test to fail This PR updates the tests to use the values 0 (unset pid), 1 (running pid), and 4194305 (never running pid, as it's max_pid+1) it also fixes a small inconsistency in reserving space for the slice of namespace paths Signed-off-by: Peter Hunt <[email protected]>
Codecov Report
@@ Coverage Diff @@
## release-1.18 #3847 +/- ##
================================================
+ Coverage 40.42% 40.43% +0.01%
================================================
Files 105 105
Lines 8628 8628
================================================
+ Hits 3488 3489 +1
+ Misses 4828 4827 -1
Partials 312 312 |
|
/test e2e_rhel |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: openshift-cherrypick-robot, saschagrunert The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: openshift-cherrypick-robot, saschagrunert The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
/lgtm |
|
/retest |
|
/retest |
|
/retest |
|
gah, we're hitting an issue where sometimes nodes don't come up on fedora: |
|
/retest |
3 similar comments
|
/retest |
|
/retest |
|
/retest |
This is an automated cherry-pick of #3839
/assign haircommander