libct/cgroups/systemd: eliminate runc/systemd race

In case it takes more than 1 second for systemd to create a unit,
startUnit() times out with a warning and then runc proceeds
(to create cgroups using fs manager and so on).

Now runc and systemd are racing, and multiple scenarios are possible.

In one such scenario, by the time runc calls systemd manager's Apply()
the unit is not yet created, the dbusConnection.SetUnitProperties()
call fails with "unit xxx.scope not found", and the whole container
start also fails.

To eliminate the race, we need to return an error in case the timeout is
hit.

To reduce the chance to fail, increase the timeout from 1 to 30 seconds,
to not error out too early on a busy/slow system (and times like 3-5
seconds are not unrealistic).

While at it, as the timeout is quite long now, make sure to not leave
a stray timer.

Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
This commit is contained in:
Kir Kolyshkin
2020-09-29 17:18:29 -07:00
parent 49d4507c77
commit 38447895a5

View File

@@ -325,6 +325,9 @@ func isUnitExists(err error) bool {
func startUnit(dbusConnection *systemdDbus.Conn, unitName string, properties []systemdDbus.Property) error {
statusChan := make(chan string, 1)
if _, err := dbusConnection.StartTransientUnit(unitName, "replace", properties, statusChan); err == nil {
timeout := time.NewTimer(30 * time.Second)
defer timeout.Stop()
select {
case s := <-statusChan:
close(statusChan)
@@ -333,8 +336,9 @@ func startUnit(dbusConnection *systemdDbus.Conn, unitName string, properties []s
dbusConnection.ResetFailedUnit(unitName)
return errors.Errorf("error creating systemd unit `%s`: got `%s`", unitName, s)
}
case <-time.After(time.Second):
logrus.Warnf("Timed out while waiting for StartTransientUnit(%s) completion signal from dbus. Continuing...", unitName)
case <-timeout.C:
dbusConnection.ResetFailedUnit(unitName)
return errors.New("Timeout waiting for systemd to create " + unitName)
}
} else if !isUnitExists(err) {
return err