>>> "Gao,Yan" <[email protected]> schrieb am 13.10.2011 um 03:55 in Nachricht
<[email protected]>:
> On 10/12/11 15:42, Ulrich Windl wrote:
> > Hello,
> >
> > I have a question: when I have a resource group where the last resource has
> utilization attributes, I see the following effect (SLES11 SP1):
> >
> > All the resources start up on one node, but not the one with utilization
> attributes. So effectively the group does not start up successfully. This is
> especially bad that another node in the cluster would have the capacity to
> start up that resource with utilization attributes.
> Could you please send out your configuration?
A sketch:
node h01 \
utilization utl_ram="1200" utl_cpu="200" \
attributes standby="off"
node h05 \
utilization utl_ram="1200" utl_cpu="200" \
attributes standby="off"
[...]
primitive prm_v01_xen ocf:heartbeat:Xen \
params xmfile="/etc/xen/vm/v01" \
op start interval="0" timeout="300" \
op stop interval="0" timeout="300" \
op monitor interval="1200" timeout="90" \
utilization utl_cpu="40" utl_ram="164"
[...]
# There are six of those Xen resources, and the utl_cpu won't be enough for all
group grp_v01 prm_01_npiv_1 prm_v01_npiv_2 prm_v01_v01_E1 prm_v01_v01_E2 [...]
prm_v01_v01_raid1 prm_v01_xen meta priority="1500" resource-stickiness="500000"
Note on the stickyness: The resource should not move once started.
I think I found the problem: LRM will add 500000 to every resource, effectively
making a score of INFINITY for stickyness, right? If so, then sorry: Once I
learned about taht I changed most resources, but seem to have overlooked these.
Regards,
Ulrich
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
