ci: Test bootupd

[cgwalters]

Don't merge.

[cgwalters]

/test all

[openshift-ci-robot]

Skipping CI for Draft Pull Request.
If you want CI signal for your change, please convert it to an actual PR.
You can still manually trigger a test run with /test all

[openshift-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: cgwalters

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 [cgwalters]

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

[cgwalters]

One thing I was investigating is whether there was bootupd version drift, but it seems we have 0.2.5 everywhere - and in any case there's not much different between 0.2.4 and 0.2.5 in this respect.

[miabbott]

For the sake of completeness, this is the error we saw in the internal RHCOS pipeline (spotted by @travier)

+ /usr/bin/bootupctl backend generate-update-metadata /
error: file /boot/efi/EFI/BOOT/fbx64.efi: No such file or directory
error: file /boot/efi/EFI/redhat/grubx64.efi: No such file or directory
error: file /boot/efi/EFI/BOOT/BOOTX64.EFI: No such file or directory
error: file /boot/efi/EFI/redhat/shimx64.efi: No such file or directory
error: file /boot/efi/EFI/redhat/mmx64.efi: No such file or directory
error: file /boot/efi/EFI/redhat/shimx64-redhat.efi: No such file or directory
error: file /boot/efi/EFI/redhat/BOOTX64.CSV: No such file or directory
error: generating metadata failed: Failed to invoke rpm -qf
error: bwrap(/usr/bin/rpmostree-postprocess-inline-0): Child process killed by signal 1

[miabbott]

CI job shows the same error noted above

[cgwalters]

OK I verified that using quay.io/coreos-assembler/coreos-assembler:v0.11.0 works. But no idea why yet.

[cgwalters]

Let's land openshift/release#18338 then we'll do a PR here to change our Dockerfile to use it.

[jlebon]

I think the issue might be #548. Something like: base rpm doesn't understand the rpmdb, so rpm -qf gets confused.

[jlebon]

Looking at https://github.com/rpm-software-management/rpm/blob/847c6f062c267c4be643be9c202141bd330a7891/lib/query.c#L466-L478 at least, it does seem like rpm -qf falls back to doing a stat if it can't query the rpmdb and then printing the error message we see in the compose because it can't find the file (which is expected during the treecompose).

[cgwalters]

/retest

[cgwalters]

OK, our CI is working again.