Skip to content

Conversation

@openshift-cherrypick-robot

This is an automated cherry-pick of #4113

/assign joelcapitao

The test needs to be adapted so it doesn't verify that we're on
multipath based on whether the device is `/dev/mapper/mpath` or not,
but instead based on udevadm output.
This verification works whether or not the `user_friendly_names`
option is used while configuring multipath.
In order to have a more generic test, let's not use user-friendly names
while configuring multipath and specify instead a World Wide ID to the
stub device. We also adapt the systemd units accordingly.
@openshift-ci
Copy link

openshift-ci bot commented Nov 4, 2025

Hi @openshift-cherrypick-robot. Thanks for your PR.

I'm waiting for a github.com 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.

@joelcapitao
Copy link
Member

/ok-to-test

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request adapts the multipath tests to handle UUID-based device names, which is a good improvement for test reliability. The changes to use mpathconf --user_friendly_names n, set a wwn for the disk, and update systemd units are all consistent and logical. The updated verification logic in verifyMultipath is also more robust. I have one suggestion to make the verification even more reliable by handling potential whitespace in command output.

Comment on lines 153 to +154
srcdev := string(c.MustSSHf(m, "findmnt -nvr %s -o SOURCE", path))
if !strings.HasPrefix(srcdev, "/dev/mapper/mpath") {
udevinfo := string(c.MustSSHf(m, "udevadm info %s", srcdev))
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The output of c.MustSSHf includes the raw standard output of the command, which might include trailing whitespace or newlines. This could cause the subsequent udevadm info command to fail or behave unexpectedly. It's safer to trim the whitespace from the srcdev string before using it in the next command.

Suggested change
srcdev := string(c.MustSSHf(m, "findmnt -nvr %s -o SOURCE", path))
if !strings.HasPrefix(srcdev, "/dev/mapper/mpath") {
udevinfo := string(c.MustSSHf(m, "udevadm info %s", srcdev))
srcdev := strings.TrimSpace(string(c.MustSSHf(m, "findmnt -nvr %s -o SOURCE", path)))
udevinfo := string(c.MustSSHf(m, "udevadm info %s", srcdev))

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request adapts multipath tests to work with non-user-friendly device names. This involves updating systemd unit files to use stable, WWID-based device paths, configuring the test to create a disk with a specific WWN, and making the verification logic more robust by checking udev properties instead of relying on device name prefixes. The changes are consistent and correct. I have one suggestion to improve security and robustness in the verification function by properly quoting a variable used in a shell command.

func verifyMultipath(c cluster.TestCluster, m platform.Machine, path string) {
srcdev := string(c.MustSSHf(m, "findmnt -nvr %s -o SOURCE", path))
if !strings.HasPrefix(srcdev, "/dev/mapper/mpath") {
udevinfo := string(c.MustSSHf(m, "udevadm info %s", srcdev))
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The srcdev variable is obtained from the output of a command and then used to construct another shell command. This could be a potential command injection vulnerability if srcdev contains shell metacharacters. It's safer to quote it. Using the %q format verb in fmt.Sprintf will properly quote the string, preventing misinterpretation by the shell.

Suggested change
udevinfo := string(c.MustSSHf(m, "udevadm info %s", srcdev))
udevinfo := string(c.MustSSHf(m, "udevadm info %q", srcdev))

@joelcapitao
Copy link
Member

/override continuous-integration/jenkins/pr-merge

For some reason the worker pod is terminated consistently when cheking format with gofmt

@openshift-ci
Copy link

openshift-ci bot commented Nov 4, 2025

@joelcapitao: Overrode contexts on behalf of joelcapitao: continuous-integration/jenkins/pr-merge

In response to this:

/override continuous-integration/jenkins/pr-merge

For some reason the worker pod is terminated consistently when cheking format with gofmt

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.

Copy link
Contributor

@yasminvalim yasminvalim left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

/LGTM

@joelcapitao joelcapitao merged commit 31009d2 into coreos:rhcos-4.18 Nov 4, 2025
6 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants