5

I'm playing around with Red Hat's High Availability Add-On, everything seems working fine however I have one issue:

Whenever I reboot one of nodes 7test2, resource group g1 migrates to 7test1 (by design) and then when 7test2 comes back online resource group g1 migrates back, it's like it has some sort of stickiness or something.

[root@7test1 ~]# cat /etc/redhat-release 
Red Hat Enterprise Linux Server release 7.1 (Maipo)
[root@7test1 ~]# rpm -q pcs 
pcs-0.9.137-13.el7.x86_64
[root@7test1 ~]# pcs status
Cluster name: 7test
Last updated: Thu Apr 23 17:11:10 2015
Last change: Thu Apr 23 17:00:02 2015
Stack: corosync
Current DC: 7test1.uftwf.local (1) - partition with quorum
Version: 1.1.12-a14efad
2 Nodes configured
3 Resources configured


Online: [ 7test1.uftwf.local 7test2.uftwf.local ]

Full list of resources:

 rhevm  (stonith:fence_rhevm):  Started 7test1.uftwf.local 
 Resource Group: g1
     IPaddr (ocf::heartbeat:IPaddr2):   Started 7test2.uftwf.local 
     MailTo (ocf::heartbeat:MailTo):    Started 7test2.uftwf.local 

PCSD Status:
  7test1.uftwf.local: Online
  7test2.uftwf.local: Online

Daemon Status:
  corosync: active/enabled
  pacemaker: active/enabled
  pcsd: active/enabled
[root@7test1 ~]# 

How can I set resource stickiness in pacemaker?

alexus
  • 12,342
  • 27
  • 115
  • 173

3 Answers3

3

You can set the value of resource-stickiness, a resource meta option that indicates how much the resource prefers to stay where it is (the default value is 0) for a given resource or group whith the following command:

# pcs resource meta resource_id resource-stickiness=100

For example:

# pcs resource meta dns_ip resource-stickiness=100

# pcs resource show dns_ip Resource: dns_ip (class=ocf provider=heartbeat type=IPaddr2) Attributes: ip=10.0.0.1 cidr_netmask=24 Meta Attrs: resource-stickiness=100 Operations: start interval=0s timeout=20s (dns_ip-start-timeout-20s)

2

To display the current value of a specific cluster property, use the following command.

pcs property list --all

To display the current value of a specific cluster property, use the following command.

pcs property show property

To set the value of a cluster property, use the following pcs command.

pcs property set property=value
c4f4t0r
  • 5,149
  • 3
  • 28
  • 41
0

In most circumstances, it is highly desirable to prevent healthy resources from being moved around the cluster. Moving resources almost always requires a period of downtime. For complex services such as databases, this period can be quite long.

To address this, Pacemaker has the concept of resource stickiness, which controls how strongly a service prefers to stay running where it is. You may like to think of it as the "cost" of any downtime. By default, Pacemaker assumes there is zero cost associated with moving resources and will do so to achieve "optimal" [6] resource placement. We can specify a different stickiness for every resource, but it is often sufficient to change the default.

⸺ from clusterlabs.org

Pang
  • 273
  • 3
  • 8
Vijay S B
  • 185
  • 1
  • 4