test: use block devices for I/O limit tests by giuseppe · Pull Request #26022 · podman-container-tools/podman

giuseppe · 2025-04-30T09:42:19Z

change the device used for block I/O limits (read/write BPS and IOPS).

Since /dev/zero is not a block device, it's unsuitable for testing cgroup controls designed for block devices.

None

packit-as-a-service · 2025-04-30T12:41:22Z

[NON-BLOCKING] Packit jobs failed. @containers/packit-build please check. Everyone else, feel free to ignore.

giuseppe · 2025-04-30T13:13:31Z

@Luap99 PTAL

Luap99

LGTM, a small hint for module loading would be nice. These tests are run in many external CI systems so it would be helpful for them to know how to enable this test without having to search for this specific kernel device.

Luap99 · 2025-04-30T14:06:37Z

 	}
 }

+func SkipIfNotExist(path string) {


can you add a reason field, it would be nice to pass a hint, i.e. run modprobe null_blk nr_devices=1 so external consumers know directly how to make the test work.

Luap99 · 2025-04-30T14:07:14Z

    skip_if_rootless "cannot use this flag in rootless mode"

+    if test \! -e /dev/nullb0; then
+        skip "/dev/nullb0 not present"


same here include the modprobe null_blk nr_devices=1 info please

Cgroup block I/O limits cannot be applied to character devices. Ignore character devices in the inspect output. Update the API tests to use the null block device `/dev/nullb0` (if available) instead of `/dev/zero` for testing I/O limits. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>

The tests were incorrectly using `/dev/zero`. These options are intended to set I/O limits on specific block devices. The test already sets up a loopback device, so reuse it. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>

The tests for device I/O limits were using `/dev/zero`, which is not a block device suitable for these cgroup controls. Update the tests to use `/dev/nullb0` if it exists. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>

Load the `null_blk` kernel module to have the `/dev/nullb0` device available for the tests. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>

Luap99

LGTM

/cherry-pick v5.5

openshift-ci · 2025-04-30T14:27:46Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: giuseppe, Luap99

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

The pull request process is described here

Details

Needs approval from an approver in each of these files:

~~OWNERS~~ [Luap99,giuseppe]

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

Luap99 · 2025-04-30T14:31:22Z

@siretart @ricardobranco777 As I know you run these tests downstream just a heads up that you may have to run modprobe null_blk nr_devices=1 on your test systems to keep coverage of this test otherwise it will get skipped.

ricardobranco777 · 2025-04-30T16:39:02Z

Since /dev/zero is not a block device, it's unsuitable for testing cgroup controls designed for block devices.

Thanks for the heads-up!

giuseppe · 2025-04-30T16:44:36Z

@mheon PTAL

mheon · 2025-04-30T18:10:29Z

/lgtm

Too late to make RC2 unfortunately, but it should be in 5.5 final

Luap99 · 2025-05-02T10:13:33Z

/cherry-pick v5.5

openshift-cherrypick-robot · 2025-05-02T10:14:31Z

@Luap99: new pull request created: #26049

Details

In response to this:

/cherry-pick v5.5

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.

Luap99 · 2025-05-02T10:14:49Z

@lsm5 @jnovy We will also need to run modprobe null_blk nr_devices=1 in the fedora/RHEL gating envs before the tests.

openshift-ci Bot added do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. release-note-none approved Indicates a PR has been approved by an approver from all required OWNERS files. labels Apr 30, 2025

giuseppe force-pushed the test-do-not-set-limits-on-dev-zero branch 5 times, most recently from f4fd259 to cfc811a Compare April 30, 2025 12:35

giuseppe force-pushed the test-do-not-set-limits-on-dev-zero branch from cfc811a to fd918cb Compare April 30, 2025 13:12

giuseppe changed the title ~~[WIP][DO NOT REVIEW] citest: Use block devices for I/O limit tests~~ test: use block devices for I/O limit Apr 30, 2025

giuseppe changed the title ~~test: use block devices for I/O limit~~ test: use block devices for I/O limit tests Apr 30, 2025

giuseppe marked this pull request as ready for review April 30, 2025 13:13

openshift-ci Bot removed the do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. label Apr 30, 2025

giuseppe force-pushed the test-do-not-set-limits-on-dev-zero branch from fd918cb to 3ec6c51 Compare April 30, 2025 14:00

Luap99 added the 5.5 label Apr 30, 2025

Luap99 reviewed Apr 30, 2025

View reviewed changes

giuseppe added 4 commits April 30, 2025 16:21

test/system: Use correct device for I/O limit tests

3b61e56

The tests were incorrectly using `/dev/zero`. These options are intended to set I/O limits on specific block devices. The test already sets up a loopback device, so reuse it. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>

test/e2e: Use nullb0 for IO limit tests

8160e02

The tests for device I/O limits were using `/dev/zero`, which is not a block device suitable for these cgroup controls. Update the tests to use `/dev/nullb0` if it exists. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>

ci: Load null_blk for I/O limit tests

80065f7

Load the `null_blk` kernel module to have the `/dev/nullb0` device available for the tests. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>

giuseppe force-pushed the test-do-not-set-limits-on-dev-zero branch from 3ec6c51 to 80065f7 Compare April 30, 2025 14:21

Luap99 approved these changes Apr 30, 2025

View reviewed changes

ricardobranco777 mentioned this pull request Apr 30, 2025

containers/bats: Load null_blk module for podman tests os-autoinst/os-autoinst-distri-opensuse#21929

Merged

openshift-ci Bot assigned mheon Apr 30, 2025

openshift-ci Bot added the lgtm Indicates that a PR is ready to be merged. label Apr 30, 2025

openshift-merge-bot Bot merged commit 8f22a0c into podman-container-tools:main Apr 30, 2025

openshift-cherrypick-robot mentioned this pull request May 2, 2025

[v5.5] test: use block devices for I/O limit tests #26049

Merged

Luap99 mentioned this pull request Jun 10, 2025

podman-update: fix block device handling #26308

Merged

stale-locking-app Bot added the locked - please file new issue/PR Assist humans wanting to comment on an old issue or PR with locked comments. label Aug 1, 2025

stale-locking-app Bot locked as resolved and limited conversation to collaborators Aug 1, 2025

Uh oh!

Conversation

giuseppe commented Apr 30, 2025

Uh oh!

packit-as-a-service Bot commented Apr 30, 2025

Uh oh!

giuseppe commented Apr 30, 2025

Uh oh!

Luap99 left a comment

Choose a reason for hiding this comment

Uh oh!

Luap99 Apr 30, 2025

Choose a reason for hiding this comment

Uh oh!

Luap99 Apr 30, 2025

Choose a reason for hiding this comment

Uh oh!

Luap99 left a comment

Choose a reason for hiding this comment

Uh oh!

openshift-ci Bot commented Apr 30, 2025

Uh oh!

Luap99 commented Apr 30, 2025

Uh oh!

ricardobranco777 commented Apr 30, 2025

Uh oh!

giuseppe commented Apr 30, 2025

Uh oh!

mheon commented Apr 30, 2025

Uh oh!

Luap99 commented May 2, 2025

Uh oh!

openshift-cherrypick-robot commented May 2, 2025

Uh oh!

Luap99 commented May 2, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants