>>> Dejan Muhamedagic <deja...@fastmail.fm> schrieb am 04.04.2011 um 13:56 in Nachricht <20110404115618.GD3553@squib>: > Hi, > > On Mon, Apr 04, 2011 at 08:32:10AM +0200, Alain.Moulle wrote: > > Hi > > > > I got a strange message about "... max_child_count (4) reached, > > postponing execution of operation stop ..." on a resource. > > > > What is the meaning of this max_child_count ? > > lrmd (the local resource manager) won't run more than this > number of resource operations in parallel. It's a mechanism to > prevent placing too much load on the node. As soon as one of the > operation finishes, the next one in the queue is started.
Hi! I wonder: If a machine has something like 12 cores, shouldn't that value depend on the number of CPUs? Regards, Ulrich > > > Is it possible to tune it ? > > How can it be tuned ? with regard with which items ? nb of resources > > configured ? anything else ? > > Yes, by setting LRMD_MAX_CHILDREN in /etc/sysconfig/pacemaker in > SUSE distributions (SLES or OpenSUSE). Don't know if init > scripts of other distributions have this option. I think you'll > need the latest Heartbeat (if you're running heartbeat and not > corosync). Note that this is a per-node setting. > > Thanks, > > Dejan > > > > > Thanks a lot. > > Alain > > _______________________________________________ > > Linux-HA mailing list > > Linux-HA@lists.linux-ha.org > > http://lists.linux-ha.org/mailman/listinfo/linux-ha > > See also: http://linux-ha.org/ReportingProblems > _______________________________________________ > Linux-HA mailing list > Linux-HA@lists.linux-ha.org > http://lists.linux-ha.org/mailman/listinfo/linux-ha > See also: http://linux-ha.org/ReportingProblems > _______________________________________________ Linux-HA mailing list Linux-HA@lists.linux-ha.org http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
