[ClusterLabs] 32 nodes pacemaker cluster setup issue

Klaus Wenninger kwenning at redhat.com
Wed May 19 05:52:30 EDT 2021

In the logs previously in this thread one could see that corosync
isn't able to switch to realtime-scheduling.
This can lead to timing issues if there is load - doesn't have to
be excessive.
Corosync had issues enabling realtime-scheduling when
e.g. CPUAccounting was enabled. But that should be solved
for quite some time. Of course new issues might have come
Thus I guess it would be helpful if you give us the Distribution
you are using, version of it, kernel-version, versions of
pacemaker/corosync and if you took the distro-builds or your
own builds.


On 5/18/21 10:18 PM, Digimer wrote:
> On 2021-05-18 1:13 p.m., S Sathish S wrote:
>> Hi Digimer/Team,
>> In our product use unicast protocols and CPU load is normal while
>> problematic timing.
>> We don’t defined corosync / totem timing values using default timing
>> till now, Please suggest what basic we need to tune this parameter based
>> on cluster nodes ?
>> [root at node1 ~]# corosync-cmapctl | grep totem.token
>> runtime.config.totem.token (u32) = 19850
>> runtime.config.totem.token_retransmit (u32) = 4726
>> runtime.config.totem.token_retransmits_before_loss_const (u32) = 4
> I have no experience with such large clusters, so I can't offer direct
> advice. Perhaps this thread is helpful?
> https://lists.clusterlabs.org/pipermail/users/2016-August/010999.html
> You might also find useful advice in 'man 5 corosync.conf'.

Klaus Wenninger

Senior Software Engineer, EMEA ENG Base Operating Systems

Red Hat

kwenning at redhat.com

Red Hat GmbH, https://de.redhat.com/ , Registered seat: Grasbrunn,
Commercial register: Amtsgericht Muenchen, HRB 153243,
Managing Directors: Charles Cachera, Brian Klemm, Laurie Krebs, Michael O'Neill

More information about the Users mailing list