zio: do no-op injections just before handing off to vdevs #17029
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
amotin
approved these changes
Feb 6, 2025
amotin
reviewed
Feb 6, 2025
The purpose of no-op is to simulate a failure between a device cache and its permanent store. We still want it to go through the queue and respond in the same way to everything else. So, inject "success" as the very last thing, and then move on to VDEV_IO_DONE to be dequeued and so any followup work can occur. Sponsored-by: Klara, Inc. Sponsored-by: Wasabi Technology, Inc. Signed-off-by: Rob Norris <[email protected]>
robn
force-pushed
the
inject-noop-late
branch
from
5e3184a to
b31bdbd
Compare
amotin
approved these changes
Feb 7, 2025
amotin
added
Status: Accepted
tonyhutter
approved these changes
Feb 8, 2025
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Outdated suggestions cannot be applied.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.
[Sponsors: Klara, Inc., Wasabi Technology, Inc.]
Motivation and Context
"no-op" injections (#16085) simply make the op succeed, entirely bypassing the vdev stages, including queuing and error assessment. This makes it useless for testing those things in conjunction with its original purpose: simulating a failure between a device cache and its permanent store.
Description
Move the injection down, making it the last thing to happen after the IO is dequeued and the device health is checked. Then, don't bypass the vdev stages anymore; instead, let it proceed to VDEV_IO_DONE where it can be properly counted and disaggregated by the queue and any followup processing can occur.
How Has This Been Tested?
Mostly internal testing with some new queue functionality. We have no specific tests for no-op in the test suite, though I've run the
zinjectsuite to success as some sort of sanity.Types of changes
Checklist:
Signed-off-by.