pass device path on publish from controller #268
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
WanzenBug
force-pushed
the
fix-caching-issues
branch
2 times, most recently
from
24254af to
bc4ab74
Compare
When the node plugin receives a publish (== mount) call, it needs to know what device needs to be mounted. Up to now, this was done by an API request to LINSTOR. With the introduction of more aggressive caching, this often meant the request would end in a 404 if multiple volumes where mounted in short succession. To fix this, we move the retrieval of the device path out of the node plugin entirely. We can simply pass the device path from the ControllerPublish() call down using the publish context. We also remove the request for the device path from the NodeExpand() call: we have to look at the local mount table so we get the device name from there. With this change, the node plugin should no longer need to request LINSTOR API outside the node/storage pool information, which is not affected by this caching issue. Signed-off-by: Moritz Wanzenböck <[email protected]>
WanzenBug
force-pushed
the
fix-caching-issues
branch
from
bc4ab74 to
90a8469
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When the node plugin receives a publish (== mount) call, it needs to know what device needs to be mounted. Up to now, this was done by an API request to LINSTOR. With the introduction of more aggressive caching, this often meant the request would end in a 404 since the cache did not know about the volume at all.
To fix this, we move the retrieval of the device path out of the node plugin entirely. We can simply pass the device path from the ControllerPublish() call down using the publish context. There, the cache is not an issue since for the volume to be created, some kind of modification is needed, which will always result in cache invalidation.
We also remove the request for the device path from the NodeExpand() call: we have to look at the local mount table so we get the device name from there.
With this change, the node plugin should no longer need to request LINSTOR API outside the node/storage pool information, which is not affected by this caching issue.