Thanks Marko - I ran your script on an up-to-date f31 VM and after it finished, I do see the "pmlogger.service failed because a timeout was exceeded" messages, however after no further action the pmlogger service is up and running:

....
Job for pmlogger.service failed because a timeout was exceeded.
See "systemctl status pmlogger.service" and "journalctl -xe" for details.


Running systemctl status pmlogger.service now shows:


[root@f31-systemd mgoodwin]# systemctl status pmlogger.service
● pmlogger.service - Performance Metrics Archive Logger
   Loaded: loaded (/usr/lib/systemd/system/pmlogger.service; enabled; vendor preset: enabled)
   Active: active (running) since Mon 2019-11-04 08:04:57 AEDT; 6s ago
     Docs: man:pmlogger(1)
  Process: 89860 ExecStart=/usr/share/pcp/lib/pmlogger start (code=exited, status=0/SUCCESS)
 Main PID: 95257 (pmlogger)
    Tasks: 1 (limit: 9506)
   Memory: 3.4M
      CPU: 6.362s
   CGroup: /system.slice/pmlogger.service
           └─95257 /usr/libexec/pcp/bin/pmlogger -P -r -T24h10m -c config.default -v 100mb -m pmlogger_check 20191104.08.04

Nov 04 08:04:52 f31-systemd systemd[1]: Starting Performance Metrics Archive Logger...
Nov 04 08:04:52 f31-systemd pmlogger[89860]: Starting pmlogger ...
Nov 04 08:04:52 f31-systemd systemd[1]: pmlogger.service: Can't open PID file /run/pcp/pmlogger.pid (yet?) after start: No such file o>
Nov 04 08:04:57 f31-systemd systemd[1]: pmlogger.service: Supervising process 95257 which is not our child. We'll most likely not noti>
Nov 04 08:04:57 f31-systemd systemd[1]: Started Performance Metrics Archive Logger.
[root@f31-systemd mgoodwin]# pcp
Performance Co-Pilot configuration on f31-systemd:

 platform: Linux f31-systemd 5.3.7-301.fc31.x86_64 #1 SMP Mon Oct 21 19:18:58 UTC 2019 x86_64
 hardware: 2 cpus, 1 disk, 1 node, 7961MB RAM
 timezone: AEDT-11
 services: pmcd pmproxy
     pmcd: Version 5.0.0-1, 9 agents, 4 clients
     pmda: root pmcd proc pmproxy xfs linux mmv kvm jbd2
 pmlogger: primary logger: /var/log/pcp/pmlogger/f31-systemd/20191104.08.04
     pmie: primary engine: /var/log/pcp/pmie/f31-systemd/pmie.log


And the journalctl -xe -u pmlogger.service log shows :


-- A start job for unit pmlogger.service has begun execution.
-- 
-- The job identifier is 51208.
Nov 04 08:02:51 f31-systemd pmlogger[82555]: Starting pmlogger ...
Nov 04 08:02:51 f31-systemd systemd[1]: pmlogger.service: Can't open PID file /run/pcp/pmlogger.pid (yet?) after start: No such file or directory
Nov 04 08:02:57 f31-systemd systemd[1]: pmlogger.service: New main PID 89164 does not belong to service, and PID file is not owned by root. Refusing.
Nov 04 08:04:51 f31-systemd systemd[1]: pmlogger.service: start operation timed out. Terminating.
Nov 04 08:04:51 f31-systemd systemd[1]: pmlogger.service: Failed with result 'timeout'.
-- Subject: Unit failed
-- Defined-By: systemd
-- Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- The unit pmlogger.service has entered the 'failed' state with result 'timeout'.
Nov 04 08:04:51 f31-systemd systemd[1]: Failed to start Performance Metrics Archive Logger.
-- Subject: A start job for unit pmlogger.service has failed
-- Defined-By: systemd
-- Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- A start job for unit pmlogger.service has finished with a failure.
-- 
-- The job identifier is 51208 and the job result is failed.
Nov 04 08:04:52 f31-systemd systemd[1]: pmlogger.service: Service RestartSec=100ms expired, scheduling restart.
Nov 04 08:04:52 f31-systemd systemd[1]: pmlogger.service: Scheduled restart job, restart counter is at 1.
-- Subject: Automatic restarting of a unit has been scheduled
-- Defined-By: systemd
-- Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Automatic restarting of the unit pmlogger.service has been scheduled, as the result for
-- the configured Restart= setting for the unit.
Nov 04 08:04:52 f31-systemd systemd[1]: Stopped Performance Metrics Archive Logger.
-- Subject: A stop job for unit pmlogger.service has finished
-- Defined-By: systemd
-- Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- A stop job for unit pmlogger.service has finished.
-- 
-- The job identifier is 51547 and the job result is done.
Nov 04 08:04:52 f31-systemd systemd[1]: Starting Performance Metrics Archive Logger...
-- Subject: A start job for unit pmlogger.service has begun execution
-- Defined-By: systemd
-- Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- A start job for unit pmlogger.service has begun execution.
-- 
-- The job identifier is 51547.
Nov 04 08:04:52 f31-systemd pmlogger[89860]: Starting pmlogger ...
Nov 04 08:04:52 f31-systemd systemd[1]: pmlogger.service: Can't open PID file /run/pcp/pmlogger.pid (yet?) after start: No such file or directory
Nov 04 08:04:57 f31-systemd systemd[1]: pmlogger.service: Supervising process 95257 which is not our child. We'll most likely not notice when it exits.
Nov 04 08:04:57 f31-systemd systemd[1]: Started Performance Metrics Archive Logger.
-- Subject: A start job for unit pmlogger.service has finished successfully
-- Defined-By: systemd
-- Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- A start job for unit pmlogger.service has finished successfully.
-- 
-- The job identifier is 51547.


So it looks like the service start initially failed with a timeout, and then the restart=always kicked in after 100ms, resulting in the service starting correctly (the second time).

Is this exactly what everyone else is seeing? i.e. do you end up with the service actually running correctly, despite the initial timeout? Or is the pmlogger service still NOT running at the completion of the script?

thanks