Skip to content

(enhancement): gatecheck - Coredump/Crash related enhancements#134

Merged
kalebskeithley merged 1 commit intonfs-ganesha:centos-cifrom
Manimaran-MM:test_06022026
Feb 10, 2026
Merged

(enhancement): gatecheck - Coredump/Crash related enhancements#134
kalebskeithley merged 1 commit intonfs-ganesha:centos-cifrom
Manimaran-MM:test_06022026

Conversation

@Manimaran-MM
Copy link

@Manimaran-MM Manimaran-MM commented Feb 10, 2026

Changes

  • enable coredump for both ceph and gpfs workflow
  • collect coredump in case of failure
  • export coredump as artifacts for better visibility
  • parallel thread on observing ganesha pid

Logs

@Manimaran-MM Manimaran-MM force-pushed the test_06022026 branch 2 times, most recently from 24256cf to 529e7c2 Compare February 10, 2026 12:04
- enable coredump for both ceph and gpfs workflow
- collect coredump in case of failure
- export coredump as artifacts for better visibility
- parallel thread on observing ganesha pid

Signed-off-by: Manimaran-MM <manim@redhat.com>
Copy link

@kalebskeithley kalebskeithley left a comment

Choose a reason for hiding this comment

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

lgtm

@kalebskeithley kalebskeithley merged commit ff5b6a6 into nfs-ganesha:centos-ci Feb 10, 2026
1 check passed
@diederichm
Copy link

I wonder if we need to preserve more data (maybe its there and I do not see it)
if there is a crash -
we would want the coredump, meta data for the coredump -
but also
the ganesha binary/build along with debuginfo and debugsource rpms - so that an analysis outside of the CI environment can be performed..

@Manimaran-MM
Copy link
Author

I wonder if we need to preserve more data (maybe its there and I do not see it) if there is a crash - we would want the coredump, meta data for the coredump - but also the ganesha binary/build along with debuginfo and debugsource rpms - so that an analysis outside of the CI environment can be performed..

@diederichm This is one such job where we capture backtrace after enabling debuginfo with the same set of rpms and capturing details of "bt full" and display it as artifacts for the individual tests

https://jenkins-nfs-ganesha.apps.ocp.cloud.ci.centos.org/view/Gatecheck%20Tests/job/gatecheck-trigger-gerrithub/861/
(Please check on the artifacts that ends with "*_backtrace.txt")

I believe this should be helpful to start with initial triage... And since this crash will be specific to the individual patchset changes(which is not merged to next branch yet), we may not require to preserve the coredump/binaries to identify the root cause 🙂

@diederichm
Copy link

right - I understand we will have the backtrace with symbols.
That is certainly better then to not have it.
Not all analysis can be done from backtrace alone - it is sometimes helpful to understand the state of Ganesha (ie mdcache content, idmap cache content) - and that can be inspected if we have the core, the binaries and deuginfo/source of the build.

@Manimaran-MM
Copy link
Author

Noted.. Thanks for the feedback @diederichm .. We will review and see how we can accomodate the same on gatecheck

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants