feat: add Longhorn V2 provisioner (backport #129) #132
Conversation
![mergify[bot]](https://avatars.githubusercontent.com/in/10562?s=60&v=4){kind=small}
Signed-off-by: Tim Serong <tserong@suse.com> (cherry picked from commit 9e0fb40)
Signed-off-by: Tim Serong <tserong@suse.com> (cherry picked from commit 58b42e7)
This allows adding extra disks for use by Longhorn's V2 data engine. I've implemented this as a separate LonghornV2Provisioner for the sake of neatness, rather than grafting it into LonghornV1Provisioner with a bunch of `if` statements, but some of the implementation (the UnProvision() and Update() functions) still call back to the LonghornV1Provisioner to avoid excessive code duplication. The current implementation theoretically supports using virtio disks, except that NDM as a whole ignores those, as they don't have WWNs. This is something we need to address separately as part of harvester/harvester#6034 Related issue: harvester/harvester#5274 Signed-off-by: Tim Serong <tserong@suse.com> (cherry picked from commit 83b58dd) # Conflicts: # pkg/controller/blockdevice/controller.go
|
Cherry-pick of 83b58dd has failed: To fix up this pull request, you can check it out locally. See documentation: https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/reviewing-changes-in-pull-requests/checking-out-pull-requests-locally |
|
This pull request is now in conflict. Could you fix it @mergify[bot]? 🙏 |
|
I've fixed the conflicts, but github is still processing the update... |
|
We should use blockdevice.go ... and minor conflict with the LVM provisioner. Just curious why there is conflict. Does that mean we will have something wrong with the master branch? |
No, I think what happened is a timing problem between merging #131 and this backport opening :-/ |
|
There's something screwy going on with this PR now. I've had this "processing updates" thing for probably >30 minutes now:
...and if you look at the commit hashes in this PR, they don't match the hashes in the underyling branch (https://github.com/harvester/node-disk-manager/commits/mergify/bp/v0.7.x/pr-129/) If this doesn't resolve soon, I might just close this PR and open another one... |
|
OK, this is getting ridiculous... |
Problem:
We need to add a provisioner for LH V2 disks
Solution:
This PR
Related Issue:
harvester/harvester#6014
harvester/harvester#5274
Test plan:
TBD
Note:
The first two commits in this PR are cherry picks from #128, because I needed the changes to the blockdevices CRD. I'll drop/rebase appropriately once that PR goes in.
This is an automatic backport of pull request #129 done by Mergify.