diff options
| author | Valentin Rothberg <rothberg@redhat.com> | 2020-06-17 14:48:19 +0200 |
|---|---|---|
| committer | Valentin Rothberg <rothberg@redhat.com> | 2020-06-17 18:52:44 +0200 |
| commit | e5c3432944245a740ed443803c654dcc9c3757f0 (patch) | |
| tree | d95dd4508d1e3f847cda3787586062acecd165e1 /pkg/systemd/generate/pods_test.go | |
| parent | f293606480669acea375c53de1e2c81044313c05 (diff) | |
| download | podman-e5c3432944245a740ed443803c654dcc9c3757f0.tar.gz podman-e5c3432944245a740ed443803c654dcc9c3757f0.tar.bz2 podman-e5c3432944245a740ed443803c654dcc9c3757f0.zip | |
generate systemd: `ExecStopPost` for all units
Add an `ExecStopPost` run even for units generated without `--new`.
Although it may seem redundant to run `container/pod stop` twice at
first glance, we really need the post run. If the main PID (i.e.,
conmon) is killed, systemd will not execute `ExecStop` but only the
post one. We made this obeservation in a customer issue and could
reproduce the behavior consistently. Hence, the post run is needed
to properly clean up when conmon is killed and it's pretty much a
NOP in all other cases.
Credits to Ulrich Obergfell for throrough and detailed analyses,
which ultimately lead to this fix.
Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
Diffstat (limited to 'pkg/systemd/generate/pods_test.go')
| -rw-r--r-- | pkg/systemd/generate/pods_test.go | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/pkg/systemd/generate/pods_test.go b/pkg/systemd/generate/pods_test.go index f7ce33a30..e12222317 100644 --- a/pkg/systemd/generate/pods_test.go +++ b/pkg/systemd/generate/pods_test.go @@ -52,6 +52,7 @@ Environment=PODMAN_SYSTEMD_UNIT=%n Restart=always ExecStart=/usr/bin/podman start jadda-jadda-infra ExecStop=/usr/bin/podman stop -t 10 jadda-jadda-infra +ExecStopPost=/usr/bin/podman stop -t 10 jadda-jadda-infra PIDFile=/var/run/containers/storage/overlay-containers/639c53578af4d84b8800b4635fa4e680ee80fd67e0e6a2d4eea48d1e3230f401/userdata/conmon.pid KillMode=none Type=forking |