[ClusterLabs] systemd: xxxx.service start request repeated too quickly

[Ken Gaillot]

On 08/04/2015 09:48 AM, Andrei Borzenkov wrote:
> On Tue, Aug 4, 2015 at 4:57 PM, Juha Heinanen <jh at tutpro.com> wrote:
>> Andrei Borzenkov writes:
>>
>>> Not sure I really understand the question. If service cannot run
>>> anyway, you can simply remove it from configuration. You can set
>>> target state to stopped. You can unmanage it. It all depends on what
>>> you are attempting to achieve.
>>
>> I want pacemaker/corosync to give up automatically after a few failed
>> restarts rather that filling the disk with megabytes of syslog entries.

For this, you can use the migration-threshold resource meta-attribute.
Set it to the number of times you want to retry before giving up and
trying a different node.

>> Also, I want to add some delay to the restart attempts so that systemd
>> does not complain about too quick restarts.
> 
> This is outside of pacemaker control. "Service respawning too rapidly"
> means systemd itself attempts to restart it. You need to modify
> service definition in systemd to either disable restart on failure
> completely and let pacemaker manage it or at least add delay before
> restarts. See man systemd.service, specifically RestartSec and Restart
> parameters.
> 
>>                                                                     Now even if the mysql
>> database would become available, the service does not start, because
>> systemd tells that restart attempt by pacemaker/corosync is too quick.