Repository navigation
backend-storage: add RWO FS support - #12629
Conversation
|
Skipping CI for Draft Pull Request. |
|
/test all |
|
/test all |
1 similar comment
|
/test all |
|
/test all |
|
/test all |
Signed-off-by: Jed Lejosne <[email protected]>
Signed-off-by: Jed Lejosne <[email protected]>
Signed-off-by: Jed Lejosne <[email protected]>
Signed-off-by: Jed Lejosne <[email protected]>
…he lane Signed-off-by: Jed Lejosne <[email protected]>
|
/test pull-kubevirt-check-tests-for-flakes |
Yeah - glancing at this PR I just missed it, you are right 😊
I was thinking about what is currently achieved with the exclude-native-ssh label but maybe @0xFelix has a better POV here about whether this makes sense. It wouldn't hurt to broadcast about this new label anyway, so that everyone knows, by i.e. adding a paragraph here : Line 3 in 91e5d78 |
|
Thank you for the patience @jean-edouard |
|
Required labels detected, running phase 2 presubmits: |
|
/retest-required |
|
@jean-edouard: The following tests failed, say
DetailsInstructions 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. |
|
/test pull-kubevirt-e2e-k8s-1.31-sig-network |
|
@jean-edouard The cache fixes are required in previous versions |
|
@jean-edouard @acardace I noticed this PR only now, I didn't expect this would be merged if there are so many disagreements on the proposal! |
|
@jean-edouard could you please open an upstream issue for https://issues.redhat.com/browse/CNV-50720 |
Wouldn't it be basically the same as #13112 ? |
It's similar, even if we fix #13112 we will still have a problem when the source and the target crash. In #13112, the controller can observe that the target pod is happily running on the target already, so it can finalize the migration. (one possible solution) in the case when the source and the target are gone, we don't know how to finalize the migration. As it is today if the source pod is gone without communicating the results to the handler we cannot tell which volume to choose when the VM starts again. |
@vladikr do you think we could solve this if it was possible also to query the migration status from the target libvirt? If now, we have 2 sources where we can get this information, then at least, if one is gone, we can always use the second one. |
|
just a reminder - we should update the user-guide regarding this. |
enp0s3
left a comment
There was a problem hiding this comment.
@jean-edouard Hi, perhaps I am missing something, but which component is responsible to sync the persistent data between the source PVC and the target PVC in case of RWO FS?
Libvirt takes care of it |
What this PR does
See kubevirt/community#314
Before this PR:
After this PR:
Fixes #
Why we need it and why it was done in this way
The following tradeoffs were made:
The following alternatives were considered:
Links to places where the discussion took place:
Special notes for your reviewer
Checklist
This checklist is not enforcing, but it's a reminder of items that could be relevant to every PR.
Approvers are expected to review this list.
Release note