new testimage and systemd-image #21356
Conversation
openshift-ci
bot
added
release-note-none
approved
edsantiago
force-pushed
the
new_testimages
branch
2 times, most recently
from
ae7fb30 to
e9e7bfb
Compare
|
Cockpit tests failed for commit e9e7bfb15fd5c6c101f589ceb5c26386c3164b3e. @martinpitt, @jelly, @mvollmer please check. |
edsantiago
force-pushed
the
new_testimages
branch
2 times, most recently
from
05a8d2d to
03d3d83
Compare
| CITEST_IMAGE = "quay.io/libpod/testimage:20240123" //nolint:revive,stylecheck | ||
| SYSTEMD_IMAGE = "quay.io/libpod/systemd-image:20240124" //nolint:revive,stylecheck |
There was a problem hiding this comment.
For future: is there a good reason for duplicatingly defining CITEST_IMAGE and others in config_TAGGED.go instead of just config.go? Maybe historically these were different? I am absolutely not going to change that in this already-too-big PR, but I'd like to consider doing so in a followup.
test/system/125-import.bats
Outdated
| # FIXME: tar < 1.35 is broken. The second tar command below | ||
| # barfs with "Skipping to next header" then "Exiting with failure status" |
There was a problem hiding this comment.
DANGER WILL ROBINSON. This really creeps me out; much like #19407 this is almost certainly a bug in tar, but the other way around: last time, old-tar worked and new-rawhide-tar was broken; now new-tar works and old-tar is broken, and all I did was change a source image. I can't tell yet if this bug was intentionally fixed in 1.35, or if it's just pure happenstance that 1.35 passes. I will be looking into this today.
| @@ -33,15 +33,12 @@ function _check_health { | |||
| } | |||
|
|
|||
| @test "podman healthcheck" { | |||
| _build_health_check_image healthcheck_i | |||
There was a problem hiding this comment.
Okay, hivemind, here's a stumper:
| before | after | |
|---|---|---|
| root | 48 | 28 |
| rootless | 30 | 21 |
Time, in seconds, for hack/bats 220 (healthcheck.bats).
- root is consistently slower than rootless
- the new code shaves 20s from root, but only 9s from rootless
- times are now closer together, but root & rootless still differ
- the word "root" does not appear in
220-healthcheck.bats. No conditional skips or extra tests AFAICT. - there is a full moon
Results are consistent across O(10) runs.
Very low priority, this is a just-curious question: any ideas?
There was a problem hiding this comment.
Rootless vs root would be different systemd daemons (the user instance vs actual PID1 systemd) which could account for some of this? I assume user instance is less busy.
| LABEL created_by="$create_script @ $create_script_rev" | ||
| LABEL created_at=$create_time_z | ||
| # Note the reinstall of tzdata is required (https://bugzilla.redhat.com/show_bug.cgi?id=1870814) |
There was a problem hiding this comment.
See the BZ for details. TL;DR tzdata is no longer in fedora-minimal at all.
edsantiago
force-pushed
the
new_testimages
branch
7 times, most recently
from
fbffd4d to
a613463
Compare
edsantiago
force-pushed
the
new_testimages
branch
from
a613463 to
6e0d58b
Compare
|
ping, I'm pretty happy with this, and it's been passing (except for the new rawhide-VM flake #21504) in my hammer-CI PR. |
|
LGTM |
edsantiago
force-pushed
the
new_testimages
branch
from
6e0d58b to
2311d70
Compare
Simply because it's been a while since the last testimage
build, and I want to confirm that our image build process
still works.
Added /home/podman/healthcheck. This saves us having to
podman-build on each healthcheck test. Removed now-
unneeded _build_health_check_image helper.
testimage: bump alpine 3.16.2 to 3.19.0
systemd-image: f38 to f39
- tzdata now requires dnf **install**, not reinstall
(this is exactly the sort of thing I was looking for)
PROBLEMS DISCOVERED:
- in e2e, fedoraMinimal is now == SYSTEMD_IMAGE. This
screws up some of the image-count tests (CACHE_IMAGES).
- "alter tarball" system test now barfs with tar < 1.35.
TODO: completely replace fedoraMinimal with SYSTEMD_IMAGE
in all tests.
Signed-off-by: Ed Santiago <[email protected]>
edsantiago
force-pushed
the
new_testimages
branch
from
2311d70 to
dbfa201
Compare
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: edsantiago, Luap99 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:
Approvers can indicate their approval by writing |
stale-locking-app
bot
added
the
locked - please file new issue/PR
Simply because it's been a while since the last testimage
build, and I want to confirm that our image build process
still works.
Added /home/podman/healthcheck. This saves us having to
podman-build on each healthcheck test. Removed now-
unneeded _build_health_check_image helper.
testimage: bump alpine 3.16.2 to 3.19.0
systemd-image: f38 to f39
(this is exactly the sort of thing I was looking for)
PROBLEMS DISCOVERED:
in e2e, fedoraMinimal is now == SYSTEMD_IMAGE. This
screws up some of the image-count tests (CACHE_IMAGES).
"alter tarball" system test now barfs with tar < 1.35.
TODO: completely replace fedoraMinimal with SYSTEMD_IMAGE in all tests.
Signed-off-by: Ed Santiago [email protected]