search

samba-in-kubernetes / samba-operator

An operator for a Samba as a service on PVCs in kubernetes
Apache License 2.0
120 stars 24 forks source link

Add necessary helper scripts for CentOS CI tests #154

Closed anoopcs9 closed 2 years ago

anoopcs9 commented 2 years ago

These scripts were being groomed under a separate samba-centosci repository. For ease of access(and more control) moving those into target repository i.e, samba-operator.

CentOS CI runs are expected to fail as we do not have these scripts in the source yet. Please find the following links for successful test runs:

Above runs were triggered manually from Jenkins after making a change to point to PR source branch.

phlogistonjohn commented 2 years ago

Because of the build failures I'm going to hold off on merging these. Let me know when you're ready to merge them. Waiting may also give some others to take a look.

phlogistonjohn commented 2 years ago

....and somehow I missed your original message, expecting build failures. Still, let me know when you're ready for merge.

anoopcs9 commented 2 years ago

....and somehow I missed your original message, expecting build failures. Still, let me know when you're ready for merge.

There is one final spurious failure(on CentOS CI) to be looked at which is while deleting rook resources after successful completion of all tests resulting in an indefinite wait(or hang to be precise). I haven't bothered yet to figure out what's going on behind the curtain since it is not at all reproducible in my testing machines. May be I can refrain from deletion of rook resources and directly destroy minikube cluster as we already successfully got past real operator integration tests or somehow cancel the deletion step subject to a defined wait time?

I'll let you know in coming days.

phlogistonjohn commented 2 years ago

....and somehow I missed your original message, expecting build failures. Still, let me know when you're ready for merge.

There is one final spurious failure(on CentOS CI) to be looked at which is while deleting rook resources after successful completion of all tests resulting in an indefinite wait(or hang to be precise). I haven't bothered yet to figure out what's going on behind the curtain since it is not at all reproducible in my testing machines. May be I can refrain from deletion of rook resources and directly destroy minikube cluster as we already successfully got past real operator integration tests or somehow cancel the deletion step subject to a defined wait time?

After the tests run, immediately destroying the vms rather than incrementally cleaning up resources is fine by me.

phlogistonjohn commented 2 years ago

Feel free to kick of an (auto)merge yourself if you're ready.

anoopcs9 commented 2 years ago

All set..here we go.