[ClusterLabs] Reg: Post the pcs colocation and ordering constraint the resource gets stopped and does not start in Rocky9

Vandhana Devi K (Nokia) vandhana.devi_k at nokia.com
Thu Nov 20 02:31:36 UTC 2025 

Hi,

The following are the rpm version in the VM

pcs-0.11.9-2.el9_6.2.x86_64
pacemaker-2.1.9-1.2.el9_6.x86_64
corosync-3.1.9-2.el9_6.x86_64
resource-agents-4.10.0-71.el9_6.11.x86_64

It is a 3 node cluster and the cas_healthcheck-clone and confd_vip are applied colocation and ordering constraints.
I have attached the configurations below.
The expectation is confd_vip should be started on the node where the cas_healthcheck-clone is running.
But we see that the confd_vip is not getting started.
The same configuration worked in Rocky8 and we the failure when migrated to Rocky9.
Can you please help what could cause this behaviour in rocky9.
Attached the tail -n 500 lines of pacemaker log for reference.


[root at cit2-ops-vm0 ~]# pcs status
Cluster name: OPS
Cluster Summary:
  * Stack: corosync (Pacemaker is running)
  * Current DC: cit2-ops-vm0.katraj.com.intra (version 2.1.9-1.2.el9_6-49aab9983) - partition with quorum
  * Last updated: Thu Nov 20 07:36:54 2025 on cit2-ops-vm0.katraj.com.intra
  * Last change:  Wed Nov 19 16:46:14 2025 by root via root on cit2-ops-vm0.katraj.com.intra
  * 3 nodes configured
  * 8 resource instances configured

Node List:
  * Online: [ cit2-ops-vm0.katraj.com.intra cit2-ops-vm1.katraj.com.intra cit2-ops-vm2.katraj.com.intra ]

Full List of Resources:
  * confd_vip   (ocf:heartbeat:IPaddr2):         Stopped
  * disco_vip   (ocf:heartbeat:IPaddr2):         Started cit2-ops-vm1.katraj.com.intra
  * Clone Set: cas_healthcheck-clone [cas_healthcheck]:
    * Started: [ cit2-ops-vm0.katraj.com.intra ]
    * Stopped: [ cit2-ops-vm1.katraj.com.intra cit2-ops-vm2.katraj.com.intra ]
  * Clone Set: disco_healthcheck-clone [disco_healthcheck]:
    * Stopped: [ cit2-ops-vm0.katraj.com.intra cit2-ops-vm1.katraj.com.intra cit2-ops-vm2.katraj.com.intra ]

Daemon Status:
  corosync: active/enabled
  pacemaker: active/enabled
  pcsd: active/enabled
[root at cit2-ops-vm0 ~]# pcs resource config
Resource: confd_vip (class=ocf provider=heartbeat type=IPaddr2)
  Attributes: confd_vip-instance_attributes
    cidr_netmask=25
    ip=100.74.213.30
    nic=eth0
  Meta Attributes: confd_vip-meta_attributes
    failure-timeout=15
  Operations:
    monitor: confd_vip-monitor-interval-5s
      interval=5s
    start: confd_vip-start-interval-0s
      interval=0s timeout=20s
    stop: confd_vip-stop-interval-0s
      interval=0s timeout=20s
Resource: disco_vip (class=ocf provider=heartbeat type=IPaddr2)
  Attributes: disco_vip-instance_attributes
    cidr_netmask=25
    ip=100.77.108.19
    nic=eth2
  Meta Attributes: disco_vip-meta_attributes
    failure-timeout=15
  Operations:
    monitor: disco_vip-monitor-interval-5s
      interval=5s
    start: disco_vip-start-interval-0s
      interval=0s timeout=20s
    stop: disco_vip-stop-interval-0s
      interval=0s timeout=20s
Clone: cas_healthcheck-clone
  Meta Attributes: cas_healthcheck-clone-meta_attributes
    clone-node-max=1
    failure-timeout=15
    globally-unique=false
    migration-threshold=3
  Resource: cas_healthcheck (class=ocf provider=nokia type=rest_healthcheck)
    Attributes: cas_healthcheck-instance_attributes
      url=http://localhost:901/api/ops/cas/status
    Meta Attributes: cas_healthcheck-meta_attributes
      failure-timeout=15
    Operations:
      monitor: cas_healthcheck-monitor-interval-1s
        interval=1s
      start: cas_healthcheck-start-interval-0s
        interval=0s timeout=1
      stop: cas_healthcheck-stop-interval-0s
        interval=0s timeout=1
Clone: disco_healthcheck-clone
  Meta Attributes: disco_healthcheck-clone-meta_attributes
    clone-node-max=1
    failure-timeout=15
    globally-unique=false
    migration-threshold=3
  Resource: disco_healthcheck (class=ocf provider=nokia type=rest_healthcheck)
    Attributes: disco_healthcheck-instance_attributes
      url=https://localhost:3245/api/internal/discovery?res=health
    Meta Attributes: disco_healthcheck-meta_attributes
      failure-timeout=15
    Operations:
      monitor: disco_healthcheck-monitor-interval-1s
        interval=1s
      start: disco_healthcheck-start-interval-0s
        interval=0s timeout=1
      stop: disco_healthcheck-stop-interval-0s
        interval=0s timeout=1
[root at cit2-ops-vm0 ~]# pcs constraint config
Colocation Constraints:
  resource 'confd_vip' with resource 'cas_healthcheck-clone'
    score=INFINITY
  resource 'disco_vip' with resource 'disco_healthcheck-clone'
    score=INFINITY
Order Constraints:
  start resource 'cas_healthcheck-clone' then start resource 'confd_vip'
    kind=Mandatory

[]# cat /opt/nokia/ha/scripts/pcmkconfig/cluster/cluster.d/ops_cluster.cfg
pcs resource create confd_vip ocf:heartbeat:IPaddr2 ip={confd_vip_ipv4} cidr_netmask={confd_vip_subnet_ipv4} nic=eth0 meta failure-timeout=15 op monitor interval=5s --disabled
pcs resource create disco_vip ocf:heartbeat:IPaddr2 ip={disco_vip_ipv4} cidr_netmask={disco_vip_subnet_ipv4} nic=eth2 meta failure-timeout=15 op monitor interval=5s
pcs resource create confd_vip_ipv6 ocf:heartbeat:IPaddr2 ip={confd_vip_ipv6} cidr_netmask={confd_vip_subnet_ipv6} nic=eth0 meta failure-timeout=15 op monitor interval=5s --disabled
pcs resource create disco_vip_ipv6 ocf:heartbeat:IPaddr2 ip={disco_vip_ipv6} cidr_netmask={disco_vip_subnet_ipv6} nic=eth2 meta failure-timeout=15 op monitor interval=5s

pcs resource create cas_healthcheck ocf:nokia:rest_healthcheck url=http://localhost:9001/api/ops/cas/status meta failure-timeout=15 op monitor interval=1s --disabled
pcs resource clone cas_healthcheck meta globally-unique=false clone-node-max=1 meta failure-timeout=15 migration-threshold=3

if confd_vip_ipv4 pcs constraint colocation add confd_vip with cas_healthcheck-clone INFINITY
if confd_vip_ipv6 pcs constraint colocation add confd_vip_ipv6 with cas_healthcheck-clone INFINITY
if confd_vip_ipv4 pcs constraint order start cas_healthcheck-clone then start confd_vip kind=Mandatory
if confd_vip_ipv6 pcs constraint order start cas_healthcheck-clone then start confd_vip_ipv6 kind=Mandatory

if confd_vip_ipv4 pcs resource enable confd_vip
if confd_vip_ipv6 pcs resource enable confd_vip_ipv6
pcs resource enable cas_healthcheck
pcs resource enable cas_healthcheck-clone

#Discovery healthcheck agent
pcs resource create disco_healthcheck ocf:nokia:rest_healthcheck url=https://localhost:32245/api/internal/discovery?res=health meta failure-timeout=15 op monitor interval=1s
pcs resource clone disco_healthcheck globally-unique=false clone-node-max=1 meta failure-timeout=15 migration-threshold=3
if disco_vip_ipv4 pcs constraint colocation add disco_vip with disco_healthcheck-clone INFINITY
if disco_vip_ipv6 pcs constraint colocation add disco_vip_ipv6 with disco_healthcheck-clone INFINITY


Regards,
K.Vandhana Devi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20251120/a641ea62/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pacemaker.log
Type: application/octet-stream
Size: 110016 bytes
Desc: pacemaker.log
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20251120/a641ea62/attachment-0001.obj>

More information about the Users mailing list