Skip to content

Conversation

@shajmakh
Copy link
Contributor

@shajmakh shajmakh commented Nov 7, 2025

For High-performance configuration, cri-o started supporting exec-cpu-affinity feature and when configured to first it provides the ability for exec process to be pinned to the first CPU from the shared-CPUs IF set or to the first one from the islolated set.
(see cri-o/cri-o@4dd7fb9)

In performance profile, we want to enable this high-performance feature by default, and disbale it (legacy) it provides an annotation option.

The annotation is there just as a backup in case of bugs getting reported by the consequences of this feature enablement, and should be removed in 2 releases.

@openshift-ci
Copy link
Contributor

openshift-ci bot commented Nov 7, 2025

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: shajmakh
Once this PR has been reviewed and has the lgtm label, please assign marsik for approval. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Details Needs approval from an approver in each of these files:

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@openshift-ci openshift-ci bot requested review from MarSik and Tal-or November 7, 2025 14:07
@shajmakh
Copy link
Contributor Author

shajmakh commented Nov 7, 2025

/hold
need to add e2e test

@openshift-ci openshift-ci bot added the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Nov 7, 2025
@shajmakh
Copy link
Contributor Author

/cc @bartwensley

@openshift-ci openshift-ci bot requested a review from bartwensley November 11, 2025 09:33
@openshift-merge-robot openshift-merge-robot added the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Nov 11, 2025
@openshift-merge-robot openshift-merge-robot removed the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Nov 11, 2025
@shajmakh
Copy link
Contributor Author

/unhold

@openshift-ci openshift-ci bot removed the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Nov 12, 2025
Copy link
Contributor

@bartwensley bartwensley left a comment

Choose a reason for hiding this comment

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

This looks good to me (once the tests are passing).

@shajmakh shajmakh force-pushed the exec-affinity-pp branch 4 times, most recently from 43944ca to 57c2a0e Compare November 13, 2025 13:08
@shajmakh
Copy link
Contributor Author

@bartwensley @Tal-or thanks for the review comments, I addressed them and extended e2e test coverage and filed a PR here for more review flexibility:
#1432

@shajmakh
Copy link
Contributor Author

/hold

@openshift-ci openshift-ci bot added the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Nov 14, 2025
@shajmakh
Copy link
Contributor Author

shajmakh commented Nov 14, 2025

failed to retrieve cgroup config for pod. pod=\"performance-addon-operators-testing/test-r4llw\", container=\"test\"; failed to run command [/bin/cat /sys/fs/cgroup/cpuset.cpus.partition]: output \"\"; error \"sched_setaffinity: Invalid argument\\n\"; command terminated with exit code 1",

filed an ocp bug on that: OCPNODE-3891

@shajmakh
Copy link
Contributor Author

/retest

@shajmakh shajmakh changed the title perfprof: enable exec-cpu-affinity by default perfprof: enable exec-cpu-affinity by default (annotation) Nov 14, 2025
For High-performance configuration, cri-o started supporting exec-cpu-affinity
feature and when configured to `first` it provides the ability for exec
process to be pinned to the first CPU from the shared-CPUs IF set
or to the first one from the islolated set.
(see cri-o/cri-o@4dd7fb9)

In performance profile, we want to enable this high-performance feature
by default, and disbale it (legacy) it provides an annotation option.

The annotation is there just as a backup in case of bugs getting
reported by the consequences of this feature enablement, and should be
removed in 2 releases.

Run `./hack/render-sync.sh` to update the (no-cluster) e2e tests
expected outputs.

Signed-off-by: Shereen Haj <shajmakh@redhat.com>
u/s CI stops the run after 2 failures of the same test, which prevents a
full run of the tests. The fail-fast option is better be removed in test
runs that do not involve node reboots, that way the CI would reflect a
full run of the suite which would help us save reruns to see the next
failing test. In other words, removing it will reflect all of the
failing tests in one run.

remove this flag initially for suites of PP that are known to not have
reboots.

Signed-off-by: Shereen Haj <shajmakh@redhat.com>
@shajmakh
Copy link
Contributor Author

/unhold
since the bugs are unrelated to PP part

@openshift-ci openshift-ci bot removed the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Nov 14, 2025
@shajmakh
Copy link
Contributor Author

shajmakh commented Nov 14, 2025

/hold
after an internal conversation, we don't want to get this in to avoid breaking current operations.

@openshift-ci openshift-ci bot added the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Nov 14, 2025
@openshift-ci
Copy link
Contributor

openshift-ci bot commented Nov 14, 2025

@shajmakh: The following tests failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name Commit Details Required Rerun command
ci/prow/e2e-hypershift-pao b27abcc link true /test e2e-hypershift-pao
ci/prow/e2e-hypershift b27abcc link true /test e2e-hypershift
ci/prow/e2e-gcp-pao b27abcc link true /test e2e-gcp-pao
ci/prow/e2e-aws-ovn b27abcc link true /test e2e-aws-ovn

Full PR test history. Your PR dashboard.

Details

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.

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

Labels

do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants