bats tests - parallelize

[edsantiago]

All bats tests run with custom root/runroot, so it should be
possible to parallelize them.

(As of this initial commit, tests fail on my laptop, and I expect them to fail here. I just want to get a sense for how things go.)

Signed-off-by: Ed Santiago santiago@redhat.com

None

[openshift-ci]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: edsantiago
Once this PR has been reviewed and has the lgtm label, please assign rhatdan for approval. For more information see the Kubernetes Code Review Process.

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

Needs approval from an approver in each of these files:

• OWNERS

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

[packit-as-a-service]

Ephemeral COPR build failed. @containers/packit-build please check.

[Luap99]

I assume once we enabled parallel runs here we can port that over for the bud tests on podman? I like to get the speed up there too.

I haven't looked deeply into the prefetch logic changes but this looks much easier than podman so that is good.

[Luap99]

I would use $(nproc) here instead of hard coding any count, should make it much faster when run locally.

[github-actions]

A friendly reminder that this PR had no activity for 30 days.

[mtrmac]

(A brief look after a link from containers/skopeo#2437 )

zstd found even though push was without --force-compression: My first guess is that the FROM alpine; _EOF - built image (via build caching?) matches an image created in some other concurrently running test, and that causes BlobInfoCache to have a record about the zstd-compressed layer version.

For this purpose, it would be better to have the image’s layers be clearly independent from any other test — maybe a FROM scratch adding a file that contains the test name + timestamp.

[mtrmac]

… oh, and: debug-level logs could help.

[edsantiago]

For this purpose, it would be better to have the image’s layers be clearly independent from any other test — maybe a FROM scratch adding a file that contains the test name + timestamp.

Thank you. I thought I had checked for conflicts, but must've missed something. I'll look into this again when time allows.

[edsantiago]

push test still flaking, and I'm giving up for the day. It is stumping me:

not ok 50 bud: build push with --force-compression

# # buildah build ...
...
# f0e6fc41f58b93d990cb24331c648bc84b066822d06782aec493d97bc2b7b263
# # buildah push [...]--compression-format gzip img-t50-nesrbrf5 docker://localhost:35311/img-t50-nesrbrf5
...
# # buildah push [...] --compression-format zstd --force-compression=false img-t50-nesrbrf5 docker://localhost:35311/img-t50-nesrbrf5
...
# # skopeo inspect img-t50-nesrbrf5
# {"schemaVersion":2,"mediaType":"application/vnd.oci.image.manifest.v1+json","config":{"mediaType":"application/vnd.oci.image.config.v1+json","digest":"sha256:f0e6fc41f58b93d990cb24331c648bc84b066822d06782aec493d97bc2b7b263","size":524},"layers":[{"mediaType":"application/vnd.oci.image.layer.v1.tar+zstd","digest":"sha256:05d36d22e1ad1534254c6965a3b43cf39f4dca9d5c95551eccf40108f076da2b","size":146}],"annotations":{"org.opencontainers.image.base.digest":"","org.opencontainers.image.base.name":""}}
# #/vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv
# #|     FAIL: zstd found even though push was without --force-compression
# #| expected: !~ 'zstd'

Where is that 05d36 layer coming from?

[mtrmac]

I can’t see any obvious reason. I’d suggest doing the pushes with --log-level=debug, that should include traces like Trying to reuse blob …

[siteshwar]

I understand this pull request is in draft state, but I got this warning on OpenScanHub:

Error: SHELLCHECK_WARNING ([CWE-398](https://cwe.mitre.org/data/definitions/398.html)): [[#def1]](https://openscanhub.fedoraproject.org/task/21458/log/added.html#def1)
/usr/share/buildah/test/system/helpers.bash:211:24: warning[[SC2115](https://github.com/koalaman/shellcheck/wiki/SC2115)]: Use "${var:?}" to ensure this never expands to / .
#  209|               else
#  210|                   # Failed. Clean up, so we don't leave incomplete remnants
#  211|->                 rm -fr $_BUILDAH_IMAGE_CACHEDIR/$fname
#  212|               fi
#  213|

It should be fixed before merging this pull request.

[edsantiago]

Got it with debug. In-page search for 10b119 demonstrates the flake.

[edsantiago]

another one

[mtrmac]

Got it with debug. In-page search for 10b119 demonstrates the flake.

Note to self:
First:

[+2477s] # time="2024-11-07T16:25:26Z" level=debug msg="Ignoring BlobInfoCache record of digest \"sha256:e4216d41a8ad46a3b75f9cdc2f40cab3cd56837931d4e73be4521a9537d7cde1\", uncompressed format does not match

OK. Then:

[+2477s] # time="2024-11-07T16:26:03Z" level=debug msg="Ignoring BlobInfoCache record of digest \"sha256:e4216d41a8ad46a3b75f9cdc2f40cab3cd56837931d4e73be4521a9537d7cde1\" with unknown compression"

How did we lose the compression format knowledge??

That’s not the immediate cause, but it suggests something unexpected is happening. Both have Using SQLite blob info cache at /var/lib/containers/cache/blob-info-cache-v1.sqlite.