Recent CI tests failures

[keszybz]

--x-- Running TEST-42-EXECSTOPPOST --x--
make: Entering directory '/tmp/autopkgtest.ZCr9be/build.piX/systemd/test/TEST-42-EXECSTOPPOST'
TEST-42-EXECSTOPPOST SETUP: test that ExecStopPost= is always run
Reusing existing cached image /tmp/autopkgtest.ZCr9be/build.piX/systemd/build-deb/test/TEST-42-EXECSTOPPOST/../default.img → /tmp/autopkgtest.ZCr9be/build.piX/systemd/build-deb/test/default.img
'/var/tmp/systemd-test.JmxCFD/default.img' -> '/tmp/autopkgtest.ZCr9be/build.piX/systemd/build-deb/test/default.img'
TEST-42-EXECSTOPPOST RUN: test that ExecStopPost= is always run
+ env --unset=UNIFIED_CGROUP_HIERARCHY --unset=SYSTEMD_NSPAWN_UNIFIED_HIERARCHY /tmp/autopkgtest.ZCr9be/build.piX/systemd/build-deb/systemd-nspawn --register=no --kill-signal=SIGKILL --directory=/var/tmp/systemd-test.JmxCFD/root --setenv=SYSTEMD_UNIT_PATH=/usr/lib/systemd/tests/testdata/testsuite-42.units:/usr/lib/systemd/tests/testdata/units: /lib/systemd/systemd systemd.unit=testsuite.target systemd.wants=testsuite-42.service systemd.wants=end.service
Spawning container root on /var/tmp/systemd-test.JmxCFD/root.
Press ^] three times within 1s to kill container.
Failed to allocate scope: Failed to activate service 'org.freedesktop.systemd1': timed out (service_start_timeout=25000ms)
E: nspawn failed with exit code 1
Spawning getter /tmp/autopkgtest.ZCr9be/build.piX/systemd/build-deb/journalctl -o export -D /var/tmp/systemd-test.JmxCFD/root/var/log/journal/*...
Failed to open /var/tmp/systemd-test.JmxCFD/root/var/log/journal/*: No such file or directory
Finishing after writing 0 entries
Failed to open /var/tmp/systemd-test.JmxCFD/root/var/log/journal/*: No such file or directory
rm: cannot remove '/var/tmp/systemd-test.JmxCFD/root/var/log/journal/*': No such file or directory---- /var/tmp/systemd-test.JmxCFD/root/var/log/journal/* ----

TEST-42-EXECSTOPPOST: (failed; see logs)

[keszybz]

Various bionic-* tests are affected. I tried to figure out when this happens for the first time, and #19882 was certainly affected. But I don't see how it could be related to the failure. It seems that something changed on the distro side instead.

[yuwata]

#19900 passes all upstream tests twice. So, the candidates are

• core: Avoid spurious realization of unit cgroups #19886 (cc815b7) (See the result of Revert "core: Avoid spurious realization of unit cgroups" #19904)
• sd-event: fix failure to exit rate limiting state #19811
• Serialization cleanup #19864 (See the result of Revert "Merge pull request #19864 from keszybz/serialization-cleanup" #19908)
• dissect: try to find partition more frequently #18851 (See the result of Revert "Merge pull request #18851 from yuwata/dissect-try-to-find-par… #19907)
• Tweak the install logic again #19870 (See the result of Revert "Merge pull request #19870 from keszybz/install-foo-again" #19906)

[yuwata]

See also https://jenkins-systemd.apps.ocp.ci.centos.org/job/upstream-centos8/1364/artifact//systemd-centos-ci/artifacts_eDY5ic/testsuite-logs-upstream.yFu/coredumpctl_collect_FAIL.log