Hi,
we (Ubuntu) got a user report about the lack of restart. The topic isn't so much the crash itself (was an OOM in this case) but that the user expected sssd trying to restart in that case. I'm not sure that would have helped too much in an OOM case, but in general it seems helpful to try to get the service back up again.

Yet OTOH I'm unsure if I'm missing other reasons that would speak against such a modification of the service files. Hence I'm filing this issue to start the discussion with the experts.

So what would your opinion be about a modification like:

--- a/src/sysv/systemd/sssd.service.in
+++ b/src/sysv/systemd/sssd.service.in
@@ -11,6 +11,7 @@ ExecStart=@sbindir@/sssd -i ${DEBUG_LOGGER}
 Type=notify
 NotifyAccess=main
 PIDFile=@pidpath@/sssd.pid
+Restart=on-failure

 [Install]
 WantedBy=multi-user.target

And the same in others of ./src/sysv/systemd/sssd-* as you consider appropriate.