In debian environment we are hitting an edge-case where older buildah
version is not compatible with newer podman version because both of them
are using different storage driver.
I.e
* Podmand defaults to native `overlay`.
* Older buildah version defaults to `vfs`.
See discussions below for more details
* containers#18510 (comment)
Co-authored-by: Ed Santiago <santiago@redhat.com>
Signed-off-by: Aditya R <arajan@redhat.com>
Ginkgo test names can have more than two levels: there can be
a nested series of Describes() before the final It(). (e.g.,
quadlet_test.go). Handle that.
Before: we just assumed that the third-or-maybe-fourth line
after a "-----" divider was the test name.
Now: examine every line after the "-----" divider, until the
first empty line. Lines with /path/to/source/file are ignored,
lines with text strings are assembled together to make anchors.
This is still imperfect but it's much better than before.
SPECIAL NOTE: in order to allow linking to timing results
in the AfterSuite, I've changed the test name from Leaf to Full.
This will now be a much longer string, and hence much less
readable, but I'm inclined to think it's more correct. Please
review carefully and lmk if I should revert.
Finally, as an unrelated add-on, add links (at top) to original
log, journal, and (if applicable) podman-remote server logs.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Fix an oops in my ginkgo-v2 handling: subtest-status was
getting re-reset back to "passed", resulting in incorrect
and misleading green titles on failed tests.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Instrument system tests in hopes of tracking down #17216,
the unlinkat-ebusy-hosed flake.
Oh, also, timestamp.awk: timestamps have always been UTC, but
add a 'Z' to make it unambiguous.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Rather than supporting a special-mode *just* for netavark/aardvark
testing [in podman CI], support testing with all the latest
`podman-next` COPR packages.
The idea here is very similar to the netavark/aardvark special mode it
replaces. Most podman-dependencies do not have the level of
comprehensive CI as exist here. This new CI-mode allows testing
upstream updates to podman-dependencies without needing to roll out
a whole new package/release for them.
Also update documentation for this new mode.
Signed-off-by: Chris Evich <cevich@redhat.com>
This magic string isn't often used and may cause confusion with future
magic-string additions. Remove it.
Signed-off-by: Chris Evich <cevich@redhat.com>
In run_podman(), display a nanosecond-level timestamp next to
each command and its output.
Because this clutters the results, teach logformatter to grok
these new timestamps, strip them, and display a more human-readable
time delta in the left-hand timestamp column. logformatter started off
as a mess and is now, well, 🤮. I'm sorry. I just hope its results
make it worthwhile.
Signed-off-by: Ed Santiago <santiago@redhat.com>
With -p, ginkgo emits timing information on the first line
after dashes. Without -p, it's at the end.
Most tests use -p. Podman-machine does not. Deal with that.
Signed-off-by: Ed Santiago <santiago@redhat.com>
There are days when I really, really, really hate GNU. Remember
when someone decided that 'head -1' would no longer work, and
that it was OK to break an infinite number of legacy production
scripts? Someone now decided that egrep/fgrep are deprecated,
and our CI logs (especially pr-should-include-tests) are now
filled with hundreds of warning lines, making it difficult
to find actual errors.
I expect that those warnings will be removed quickly after
furious community backlash, just like the 'head -1' fiasco
was quietly reverted, but ITM the warnings are annoying
so I capitulate.
Signed-off-by: Ed Santiago <santiago@redhat.com>
It looks like tell/seek don't work in CI-land: important test log
information is being lost. Revert this commit, maybe some day I'll
come up with a better solution.
This reverts commit 1bff0108f6.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Super-complicated and ugly code merely to make in-page anchors
link three lines above the subtest name, to the subtest status
and timing line.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Checkpoint as of Thursday April 13 afternoon.
Please do not merge with this; the .t file (tests) needs to
be updated, and there are probably many more fixes/improvements
I still need to make.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Porting them over to v2 requires a full rewrite.
IT is not clear who actually uses these benchmarks, Valentin who wrote
them originally is in favor of removing them. He recommends to use
script from hack/perf instead.
This commit also drop the CI integration, it is not clear who actually
uses this data. If it is needed for something please speak up.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
In setup, write a containers.conf.d file with db_backend
as specified in .cirrus.yml.
This is actually much scarier and more achy-breaky than
merely "sqlite system tests": it enables sqlite in e2e
tests. ("But wait, we already do that!" -- no, not really.
sqlite in e2e is being done via --db-backend option, and
some podman commands in e2e do not use the standard options.
See #17904.
This is unlikely to get merged any time soon (March, maybe
even April) because sqlite is still too fragile; this will
trigger more flakes than are currently acceptable. Also,
the nasty auto-update flake seems to trigger much more
reliably with sqlite. We need that one fixed.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Debian system tests failing due to unwanted warning:
skipping device /dev/char/10:200 for systemd:
stat /sys/dev/char/10:200: no such file or directory
Let's see if modprobing tun will eliminate the warning.
Add loud comments requesting removal once runc bug is fixed.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Principally because 'make completion' fails if we transitively
bring in a new cobra, but also, none of the other tests are
meaningful under the treadmill.
Signed-off-by: Ed Santiago <santiago@redhat.com>
MSI Validation will fail if the process env is > 32k chars
Remove CIRRUS_COMMIT_MESSAGE and CIRRUS_PR_BODY which can easily exceed this limit
[NO NEW TESTS NEEDED]
Signed-off-by: Jason T. Greene <jason.greene@redhat.com>
The os.IsNotExist() function comment mentions that new code should use
`errors.Is(err, fs.ErrNotExist)` instead.
The check was already in CI but used the wrong function name (extra s.)
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Make sure we can install podman even when the podman-mac-helper install
command fails. This used to be the behavior but commit bae07b6ea2 caused
the regression because the binary now returns 1 as exit code on errors.
[NO NEW TESTS NEEDED] I am not sure if we can test the install step in
CI.
Fixes#17910
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
* There's a new --db-backend option. Hide it, and actually
just assume that any new --foo-backend option will
have an arg that needs to be hidden.
* Friendly-ize "Running: quadlet ..." because those messages
include a "with QUADLET_FOO=bar" string that looks like
it's part of the command line but is not, and it's really
hard for a human eye to see what's command line and what's
extra comment.
Add tests.
Signed-off-by: Ed Santiago <santiago@redhat.com>
"Just this once", I thought, adding an obscure unmaintainable
duplication to the CI code. When will I learn?
This fixes the logformatter output filename to handle boltdb/sqlite.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Future work will present podman-machine benchmark data in some useful
format for analysis. However, this data is currently only stored as CI
artifacts. When CI runs on the main branch, after a PR merges, utilize
a pair of purpose-built containers to retrieve then upload the data into
a GCE firestore database. This operation should not be critical, such
that any faults will not cause the entire CI build to be marked as a
failure.
Signed-off-by: Chris Evich <cevich@redhat.com>
Transient mode means the DB should not persist, so instead of
using the GraphRoot we should use the RunRoot instead.
Signed-off-by: Matt Heon <mheon@redhat.com>
As suggested by @edsantiago, the complex script and direct-link in the
docs are adding little value while increasing the CI maintenance burden.
Simply retire the script and strip the direct (versioned) links from the
docs.
Signed-off-by: Chris Evich <cevich@redhat.com>
The script is broken and the maintainer temporarily unavailable to help
fix it. Disable it for now, w/ reference to possible fix.
Signed-off-by: Chris Evich <cevich@redhat.com>
Also remove disused `gitlab` test setup. This test was disabled a
while ago and is unlikely to ever be revived.
Signed-off-by: Chris Evich <cevich@redhat.com>
I can't find instances of the "lookup cdn03.quay.io" flake
in RHEL gating tests. RHEL does not use systemd-resolved.
Maybe pure coincidence. Maybe not.
Let's try disabling systemd-resolved and see how things go.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Replace existing tab indentations with spaces, and add
a test to CI to prevent new ones from sneaking in.
Signed-off-by: Ed Santiago <santiago@redhat.com>
The `benchmarks.env` file is intended for machine consumption.
Including things like a `kB` unit label (like `$MEMTOTAL`) make items
difficult to parse. Additionally, multi-value keys (like `$UNAME_RM`)
make extra/unnecessary work for the interpreter. Simplify these items
and include a data-schema version marker so an interpreter can be made
aware/support future format changes.
Signed-off-by: Chris Evich <cevich@redhat.com>
The `localbenchmarks()` function stores a `.env` file containing current
environment variables for benchmark-classification purposes. However its
naked use of `printenv` means it was logging the contents of library
functions and (worse) trying to stort all the lines. This results in an
unusable mess inside `benchmarks.env`. Fix this by re-using the
purpose-built passthrough_envars() which is designed to only print
useful, safe, env. vars.
Signed-off-by: Chris Evich <cevich@redhat.com>
A semantic change to a Cirrus-CI GraphQL API parameter caused a
unit-test to fail (as it should have) with the error:
```
Query result did not pass filter '.data.ownerRepository.cronSettings':
'{"data":{"ownerRepository":null}}'
```
As per Cirrus-support, a change was introduced in schema affecting certain
fields that were incorrectly marked Nullable. They indicated the `platform`
field was set incorrectly, and should use the value `github`.
* Fix the platform field's value to `github` instead of `LINUX`.
* Change the unit-test to only execute as part of the 'main' cirrus-cron
job so it cannot impact PRs.
Signed-off-by: Chris Evich <cevich@redhat.com>
All PRs are broken because this test fails, presumably due to a GraphQL
API change. Disable the test temporarily until a permanent solution
can be found.
Signed-off-by: Chris Evich <cevich@redhat.com>
Image content hasn't changed much, the biggest thing here is the
$IMAGE_SUFFIX value. This new schema is also fully manageable by
renovate. Allowing a tag-push to c/automation_images to create image
update PRs in all repos automatically.
ref: https://github.com/containers/automation_images/pull/247
Also, cleanup a few comments and remove a disused testing task.
Signed-off-by: Chris Evich <cevich@redhat.com>
Updated build scripts and installer build scripts to include gvproxy.exe.
Includes tutorial on setting up a Podman VM with QEMU and gvproxy on Windows.
Signed-off-by: Arthur Sengileyev <arthur.sengileyev@gmail.com>
Aditya R
Ed Santiago
Lokesh Mandvekar
OpenShift Merge Robot
Anjan Nath
Chris Evich
Paul Holzinger
Jason T. Greene
Matt Heon
Erik Sjölund
Daniel J Walsh
Arthur Sengileyev