mgr/k8sevents: pin the kubernetes version to 11.0.0

[nizamial09]

There's an API change (kubernetes-client/python@c04e2f6#diff-64d3d8da58fddca2ed360257976a6ab3e28fdc4805e3480d99af857ce914dc20R27-R32) happened during last night (morning depending on your timezone) which updated the V1Events to CoreV1Events. That causes a
make check failure and this commit intends to fix it.

kubernetes-client API Change PR: kubernetes-client/python#1597

Fixes: https://tracker.ceph.com/issues/53044
Signed-off-by: Nizamudeen A [email protected]

Checklist

• References tracker ticket
• Updates documentation if necessary
• Includes tests for new functionality or reproducer for bug

---

Show available Jenkins commands

• jenkins retest this please
• jenkins test classic perf
• jenkins test crimson perf
• jenkins test signed
• jenkins test make check
• jenkins test make check arm64
• jenkins test submodules
• jenkins test dashboard
• jenkins test dashboard cephadm
• jenkins test api
• jenkins test docs
• jenkins render docs
• jenkins test ceph-volume all
• jenkins test ceph-volume tox

[nizamial09]

The first run made the tests pass although some other tests failed. (https://jenkins.ceph.com/job/ceph-pull-requests/84498/console)

243/252 Test   #4: run-tox-mgr ...............................   Passed  288.51 sec

[nizamial09]

@pcuzner @epuertat I am not entirely sure if this is correct approach. Some feedback would be nice. Also I am interested to know how we are using the kubernetes-client. Like do we get the latest version or do we pin a version? I'm assuming we are getting the latest one!

[alfonsomthd]

@jmolmo @sebastian-philipp Can you take a look at this fix? See the original error.

[sebastian-philipp]

we have to pin the version to the one shipped by our container image base (centos 8.?) Everything is not going to work.

[nizamial09]

we have to pin the version to the one shipped by our container image base (centos 8.?) Everything is not going to work.

Yes, that should be the correct way rather than expecting these kinds of errors in every API change. I am not sure how to pin the version or which version we were using before. @sebastian-philipp

[nizamial09]

Also i am not seeing the error happening on pacific, so it is using on older api.

[sebastian-philipp]

should be python3-kubernetes-1:11.0.0-6.el8.noarch according to https://jenkins.ceph.com/job/ceph-dev-build/ARCH=x86_64,AVAILABLE_ARCH=x86_64,AVAILABLE_DIST=centos8,DIST=centos8,MACHINE_SIZE=gigantic/48491//consoleFull

[sebastian-philipp]

for pacific as well: https://jenkins.ceph.com/job/ceph-dev-build/ARCH=x86_64,AVAILABLE_ARCH=x86_64,AVAILABLE_DIST=centos8,DIST=centos8,MACHINE_SIZE=gigantic/48487//consoleFull

[nizamial09]

@sebastian-philipp I think we just need to pin the kubernetes version over here right? Probably to not the latest one? https://github.com/ceph/ceph/blob/pacific/src/pybind/mgr/requirements.txt#L3

https://pypi.org/project/kubernetes/#history

[sebastian-philipp]

I'd pin it to 11.0.0 if that exist in pypi

[nizamial09]

I'd pin it to 11.0.0 if that exist in pypi

It exists, and pinned it. Updated the PR.

[ljflores]

Thank you @nizamial09!! I was looking into this last night and couldn't figure out the cause. This is very helpful.

[nizamial09]

You're welcome! @ljflores