qdiskd fails to start in RHEL 6.2 and later with "Quorum disk timings are too slow for configured token timeout" message
Issue
- When
quorumd's tkois specified in /etc/cluster/cluster.conf but notinterval,qdiskdfails to start with the message:
May 30 09:29:45 node1 qdiskd[3021]: Quorum disk timings are too slow for configured token timeout
May 30 09:29:45 node1 qdiskd[3021]: * Totem Token timeout: 10000ms
May 30 09:29:45 node1 qdiskd[3021]: * Min. Master recovery time: 18000ms
May 30 09:29:45 node1 qdiskd[3021]: Please set token timeout to at least 20000ms
Even if I raise token further it fails in the same manner.
* Cluster which has qdisk output: Quorum disk timings are too slow for configured token timeout
Environment
- Red Hat Enterprise Linux (RHEL) 6 with the High Availability Add On
- Quorum Device (qdisk) configured in
/etc/cluster/cluster.conf - One of the following attributes defined in the
quorumdtag in/etc/cluster/cluster.conf:- tko
- tko_up
- master_wait
- upgrade_wait
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
