[ClusterLabs] Q: "Multiple attributes match name=target-role"

[Xin Liang]

Hi Ulrich,

 Yes, I remembered you mentioned this multi target-role issue
https://github.com/ClusterLabs/crmsh/issues/720
________________________________
From: Users <users-bounces at clusterlabs.org> on behalf of Ulrich Windl <Ulrich.Windl at rz.uni-regensburg.de>
Sent: Thursday, June 2, 2022 2:08 PM
To: users at clusterlabs.org <users at clusterlabs.org>
Subject: [ClusterLabs] Q: "Multiple attributes match name=target-role"

Hi!

We had some issue with a resource in SLES15 SP3, so I cleaned the error:
h16:~ # crm_resource -C -r prm_xen_v07 -N h18 -n start
Cleaned up prm_xen_v07 on h18
Multiple attributes match name=target-role
  Value: Started        (id=prm_xen_v07-meta_attributes-target-role)
  Value: Started        (id=prm_xen_v07-meta_attributes-0-target-role)
  Value: Started        (id=prm_xen_v07-meta_attributes-1-target-role)

There have always been multiple target roles in the past, but I never saw this complaint, making me woner whether a recent update introduced that feature.

crmsh-4.3.1+20220505.cf4ab649-150200.5.80.1.noarch
pacemaker-2.0.5+20201202.ba59be712-150300.4.21.1.x86_64

primitive prm_xen_v07 VirtualDomain \
        params config="/etc/xen/vm/v07.xml" hypervisor="xen:///system" remoteuri="xen+tls://%n.uni-regensburg.de" autoset_utilization_cpu=false autoset_utilization_hv_memory=false \
        op start timeout=120 interval=0 \
        op stop timeout=240 interval=0 \
        op monitor interval=600 timeout=90 \
        op migrate_to timeout=120 interval=0 \
        op migrate_from timeout=120 interval=0 \
        utilization utl_cpu=20 utl_ram=2048 \
        meta priority=8900 allow-migrate=true target-role=Started \
        meta 1: resource-stickiness=5 target-role=Started \
        meta 2: rule 0: date spec hours=7-19 weekdays=1-5 resource-stickiness=1000 target-role=Started

AFAIR a crm resource start/stop set the target-role in every rule in the past.

Finally: The resource is running fine on h16, but pacemaker keeps complaining it isn't started.
Jun 02 08:03:21 h16 pacemaker-schedulerd[7494]:  warning: Unexpected result (error: Failed to start virtual domain v07.) was recorded for start of prm_xen_v07 on h19 at Jun  1 09:19:46 2022
Jun 02 08:03:21 h16 pacemaker-schedulerd[7494]:  warning: Forcing prm_xen_v07 away from h19 after 1000000 failures (max=1000000)

Also it does not try to stop it there. Could it be that this is another effect of the RAM corruption introduced with SLES15 SP3? h16 is DC...

Regards,
Ulrich



_______________________________________________
Manage your subscription:
https://lists.clusterlabs.org/mailman/listinfo/users

ClusterLabs home: https://www.clusterlabs.org/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20220602/2ec30156/attachment.htm>