[ClusterLabs] systemd dependencies

[Klaus Wenninger]

On 03/11/2019 08:12 PM, Ken Gaillot wrote:
> On Mon, 2019-03-11 at 18:18 +0000, lejeczek wrote:
>> hi guys
>> I have a pacemaker which reports:
>>
>> $ systemctl status -l pacemaker
>> ● pacemaker.service - Pacemaker High Availability Cluster 
>> Manager
>>     Loaded: loaded (/etc/systemd/system/pacemaker.service; 
>> enabled; vendor preset: disabled)
>>    Drop-In: /etc/systemd/system/pacemaker.service.d
>>             └─override.conf
>>     Active: inactive (dead)
>>       Docs: man:pacemakerd
>>
> http://clusterlabs.org/doc/en-US/Pacemaker/1.1-pcs/html/Pacemaker_Explained/index.html
>> Mar 11 10:47:41 rider.private.ccnr.ceb.private.cam.ac.uk 
>> systemd[1]: Dependency failed for Pacemaker High 
>> Availability Cluster Manager.
>> Mar 11 10:47:41 rider.private.ccnr.ceb.private.cam.ac.uk 
>> systemd[1]: Job pacemaker.service/start failed with result 
>> 'dependency'.
>>
>>
>> But corosync is up & running okey.
>> What other dependencies there are which pacemaker relies on?
>> Further more I have had this in my pacemaker service:
>>
>> [Service]
>> TimeoutStopSec=5min
>> Restart=on-failure
>> RestartSec=60
>>
>> which does not seem to help when service is in a state such 
>> as the above.
>> I can manually restart pacemaker services and it's fine.
>>
>> many thanks, L.
> That's odd. corosync is the only required dependency. There are
> optional dependencies on dbus, syslog/rsyslog, and the usually empty
> resource-agents-deps target.
Checking for 'RequiredBy=pacemaker.service' in other
unit-files might still make sense.
One of those is e.g. sbd so that if sbd is enabled pacemaker
would wait for it to start.

Klaus