btrfs recalim on kernel v5.19+: use bg_reclaim_threshold

[motiejus]

Add a sysfs knob btrfs-allocation-{,meta}data-bg_reclaim_threshold, which will do the equivalent of:

echo VALUE > /sys/fs/btrfs/FS-UUID/allocation/data/bg_reclaim_threshold

Or, in case of metadata, equivalently:

echo VALUE > /sys/fs/btrfs/FS-UUID/allocation/metadata/bg_reclaim_threshold

Where VALUE is a number between 0 and 99 inclusive.

Adding it as a "special" mount option, similarly to read_ahead_kb, as that's quite convenient.

Some resources about bg_reclaim_threshold and more broadly balancing of the btrfs filesystem:

• https://btrfs.readthedocs.io/en/latest/Administration.html#uuid-allocations-data-metadata-system
• https://web.git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=18bb8bbf13c1839b43c9e09e76d397b753989af2
• https://lwn.net/Articles/978826/ Linux v6.11+, this may obsolete bg_reclaim_threshold.

Author's interpretation

The higher the reclaim threshold, the more accurately btrfs will show unused space (Device Unallocated row of btrfs filesystem usage) at the expense of sometimes needlessly moving data around. The lower the threshold, the less rebalancing, the less accurate metrics of the remaining space.

The author of this commit prefers more IO in order to see more accurate Device Unallocated metrics, and therefore sets
btrfs-allocation-data-bg_reclaim_threshold=90.

What type of PR is this?
/kind feature

What this PR does / why we need it:

More control to the operator to manage btrfs free space tracking.

Which issue(s) this PR fixes:

Fixes #2088

Does this PR introduce a user-facing change?:

`btrfs` filesystems gain special mount options `btrfs-allocation-(data|metadata)-bg_reclaim_threshold`. The driver will write the provided value to `/sys/fs/btrfs/FS-UUID/allocation/(data|metadata)/bg_reclaim_threshold` respectively.

[k8s-ci-robot]

Hi @motiejus. Thanks for your PR.

I'm waiting for a kubernetes-sigs member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

Instructions 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.

[mattcary]

/ok-to-test

[mattcary]

Looks good to me, just a minor nit.

[mattcary]

/retest

Looks like VM failed to come up for e2e test, this is probably a flake.

[k8s-ci-robot]

@motiejus: The following test failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
pull-gcp-compute-persistent-disk-csi-driver-e2e-windows-2022	df6a569	link	false	/test pull-gcp-compute-persistent-disk-csi-driver-e2e-windows-2022

Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR.

Instructions 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.

[motiejus]

I noticed the commit has Fixes #2088, which the linter is not overly happy about.

Before I fix that and force-push, is there anything more that needs to be addressed?

[mattcary]

Hmm, it seems the linter is unhappy with the commit message, not the PR, but I can't figure out why.

Anyway

/lgtm
/approve

thank you!

I'll try to keep watch & re-lgtm if you have to force push something. Ping me if I'm slow.

[motiejus]

There is a failing Windows check — should I try to force-push and hope it's a flake that will self-resolve, or you will do something on your end?

Just checking about the next steps.

[mattcary]

I can force merge if necessary. Can you try to clear the invalid commit message first?

[motiejus]

I can force merge if necessary. Can you try to clear the invalid commit message first?

Ah, sure, done. Please kick off the tests, hopefully all will pass this time 🤞

New changes are detected. LGTM label has been removed.

This one's weird, I only changed the commit message (removed Fixes line from the bottom):

$ git diff HEAD df6a5696fcc6
$

[mattcary]

/ok-to-test

[mattcary]

I can force merge if necessary. Can you try to clear the invalid commit message first?

Ah, sure, done. Please kick off the tests, hopefully all will pass this time 🤞

New changes are detected. LGTM label has been removed.

This one's weird, I only changed the commit message (removed Fixes line from the bottom):

$ git diff HEAD df6a5696fcc6
$

Who can question the wisdom of github AI overlords?

[mattcary]

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: mattcary, motiejus

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [mattcary]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment