cluster resource for pacemaker
There is a cluster by pacemaker/corosync. The cluster consist of 3 node and doesn't have any cluster resources.
Let's try to set dummy resource.
Before load crm file, initial status is below
[root@vm01 ~]# crm_mon -1fA Stack: corosync Current DC: vm03.localdomain (version 1.1.21-1.el7-f14e36f) - partition with quorum Last updated: Wed Apr 29 17:10:15 2020 Last change: Wed Apr 29 16:58:02 2020 by hacluster via crmd on vm03.localdomain 3 nodes configured 0 resources configured Online: [ vm01.localdomain vm02.localdomain vm03.localdomain ] No active resources Node Attributes: * Node vm01.localdomain: * Node vm02.localdomain: * Node vm03.localdomain: Migration Summary: * Node vm03.localdomain: * Node vm02.localdomain: * Node vm01.localdomain: [root@vm01 ~]#
This is crm configuration file.
[root@vm01 ~]# cat dummy.crm
### Cluster Option ###
property stonith-enabled="false"
### Resource Defaults ###
rsc_defaults resource-stickiness="INFINITY" \
migration-threshold="1"
### Group Configuration ###
group grp \
resource1 \
resource2
### Clone Configuration ###
clone clnResource \
resource3
### Primitive Configuration ###
primitive resource1 ocf:heartbeat:Dummy \
op start interval="0s" timeout="300s" on-fail="restart" \
op monitor interval="10s" timeout="60s" on-fail="restart" \
op stop interval="0s" timeout="300s" on-fail="block"
primitive resource2 ocf:heartbeat:Dummy \
op start interval="0s" timeout="300s" on-fail="restart" \
op monitor interval="10s" timeout="60s" on-fail="restart" \
op stop interval="0s" timeout="300s" on-fail="block"
primitive resource3 ocf:heartbeat:Dummy \
op start interval="0s" timeout="300s" on-fail="restart" \
op monitor interval="10s" timeout="60s" on-fail="restart" \
op stop interval="0s" timeout="300s" on-fail="block"
### Resource Location ###
location rsc_location-1 grp \
rule 300: #uname eq vm01.localdomain \
rule 200: #uname eq vm02.localdomain \
rule 100: #uname eq vm03.localdomain
### Resource Colocation ###
colocation rsc_colocation-1 INFINITY: grp clnResource
### Resource Order ###
order rsc_order-1 0: clnResource grp symmetrical=false
[root@vm01 ~]#
Load crm configuration file
[root@vm01 ~]# crm configure load update dummy.crm [root@vm01 ~]#
After load crm file, Let's check.
[root@vm01 ~]# crm_mon -1fA
Stack: corosync
Current DC: vm03.localdomain (version 1.1.21-1.el7-f14e36f) - partition with quorum
Last updated: Wed Apr 29 17:10:30 2020
Last change: Wed Apr 29 17:10:27 2020 by root via cibadmin on vm01.localdomain
3 nodes configured
5 resources configured
Online: [ vm01.localdomain vm02.localdomain vm03.localdomain ]
Active resources:
Resource Group: grp
resource1 (ocf::heartbeat:Dummy): Started vm01.localdomain
resource2 (ocf::heartbeat:Dummy): Started vm01.localdomain
Clone Set: clnResource [resource3]
Started: [ vm01.localdomain vm02.localdomain vm03.localdomain ]
Node Attributes:
* Node vm01.localdomain:
* Node vm02.localdomain:
* Node vm03.localdomain:
Migration Summary:
* Node vm03.localdomain:
* Node vm02.localdomain:
* Node vm01.localdomain:
[root@vm01 ~]#
and confirm configuration by "crm configure show" command.
[root@vm01 ~]# crm configure show
node 1: vm01.localdomain
node 2: vm02.localdomain
node 3: vm03.localdomain
### Primitive Configuration ###
primitive resource1 Dummy \
op start interval=0s timeout=300s on-fail=restart \
op monitor interval=10s timeout=60s on-fail=restart \
op stop interval=0s timeout=300s on-fail=block
primitive resource2 Dummy \
op start interval=0s timeout=300s on-fail=restart \
op monitor interval=10s timeout=60s on-fail=restart \
op stop interval=0s timeout=300s on-fail=block
primitive resource3 Dummy \
op start interval=0s timeout=300s on-fail=restart \
op monitor interval=10s timeout=60s on-fail=restart \
op stop interval=0s timeout=300s on-fail=block
### Group Configuration ###
group grp resource1 resource2
### Clone Configuration ###
clone clnResource resource3
### Resource Location ###
location rsc_location-1 grp \
rule 300: #uname eq vm01.localdomain \
rule 200: #uname eq vm02.localdomain \
rule 100: #uname eq vm03.localdomain
### Resource Colocation ###
colocation rsc_colocation-1 inf: grp clnResource
### Resource Order ###
order rsc_order-1 0: clnResource grp symmetrical=false
property cib-bootstrap-options: \
have-watchdog=false \
dc-version=1.1.21-1.el7-f14e36f \
cluster-infrastructure=corosync \
stonith-enabled=false
### Resource Defaults ###
rsc_defaults rsc-options: \
resource-stickiness=INFINITY \
migration-threshold=1
[root@vm01 ~]#
try failover
delete a file to occur failure of cluster resouce as followings.
[root@vm01 ~]# rm -f /var/run/resource-agents/Dummy-resource1.state [root@vm01 ~]#
We can found failed action.
[root@vm01 ~]# crm_mon -f1A
Stack: corosync
Current DC: vm03.localdomain (version 1.1.21-1.el7-f14e36f) - partition with quorum
Last updated: Wed Apr 29 17:27:35 2020
Last change: Wed Apr 29 17:10:27 2020 by root via cibadmin on vm01.localdomain
3 nodes configured
5 resources configured
Online: [ vm01.localdomain vm02.localdomain vm03.localdomain ]
Active resources:
Resource Group: grp
resource1 (ocf::heartbeat:Dummy): Started vm02.localdomain
resource2 (ocf::heartbeat:Dummy): Started vm02.localdomain
Clone Set: clnResource [resource3]
Started: [ vm01.localdomain vm02.localdomain vm03.localdomain ]
Node Attributes:
* Node vm01.localdomain:
* Node vm02.localdomain:
* Node vm03.localdomain:
Migration Summary:
* Node vm03.localdomain:
* Node vm02.localdomain:
* Node vm01.localdomain:
resource1: migration-threshold=1 fail-count=1 last-failure='Wed Apr 29 17:27:19 2020'
Failed Resource Actions:
* resource1_monitor_10000 on vm01.localdomain 'not running' (7): call=18, status=complete, exitreason='No process state file found',
last-rc-change='Wed Apr 29 17:27:19 2020', queued=0ms, exec=0ms
[root@vm01 ~]#
A point is "Failed Resource Actions".
Migration Summary:
* Node vm03.localdomain:
* Node vm02.localdomain:
* Node vm01.localdomain:
resource1: migration-threshold=1 fail-count=1 last-failure='Wed Apr 29 17:27:19 2020'
Failed Resource Actions:
* resource1_monitor_10000 on vm01.localdomain 'not running' (7): call=18, status=complete, exitreason='No process state file found',
last-rc-change='Wed Apr 29 17:27:19 2020', queued=0ms, exec=0ms
[root@vm01 ~]#
check fail-count for resource1 -> but value=0
[root@vm01 ~]# crm resource failcount resource1 show vm01.localdomain scope=status name=fail-count-resource1 value=0 [root@vm01 ~]#
try to cleanup -> OK.
[root@vm01 ~]# crm resource cleanup resource1 vm01.localdomain
Cleaned up resource1 on vm01.localdomain
.Cleaned up resource2 on vm01.localdomain
Waiting for 1 reply from the CRMd. OK
[root@vm01 ~]#
[root@vm01 ~]# crm_mon -f1A
Stack: corosync
Current DC: vm03.localdomain (version 1.1.21-1.el7-f14e36f) - partition with quorum
Last updated: Wed Apr 29 17:43:27 2020
Last change: Wed Apr 29 17:43:17 2020 by hacluster via crmd on vm01.localdomain
3 nodes configured
5 resources configured
Online: [ vm01.localdomain vm02.localdomain vm03.localdomain ]
Active resources:
Resource Group: grp
resource1 (ocf::heartbeat:Dummy): Started vm01.localdomain
resource2 (ocf::heartbeat:Dummy): Started vm01.localdomain
Clone Set: clnResource [resource3]
Started: [ vm01.localdomain vm02.localdomain vm03.localdomain ]
Node Attributes:
* Node vm01.localdomain:
* Node vm02.localdomain:
* Node vm03.localdomain:
Migration Summary:
* Node vm03.localdomain:
* Node vm02.localdomain:
* Node vm01.localdomain:
[root@vm01 ~]#
That's all.
