cluster service failback does not work or services won't start with a failoverdomain that includes '<failoverdomainode/>' elements corresponding to the nodes in a RHEL 5 or 6 High Availability cluster with rgmanager

Solution Unverified - Updated -

Issue

  • Cluster service failback is not working. When a node starts rgmanager, the other node does not relocate the service back to it even though nofailback=0 and the starting node has a lower priority score
Aug  6 13:03:28 node2 clurgmgrd[19384]: <debug> Evaluating RG service:myApp, state started, owner node2
Aug  6 13:03:28 node2 clurgmgrd[19384]: <debug> Event (0:1:0) Processed
  • I have a restricted failoverdomain with both of my cluster nodes, and when they start rgmanager the service does not get started, and there's an error saying "Restricted domain unavailable".
Jul 24 14:34:42 node1 clurgmgrd[32761]: <notice> Marking service:myApp as stopped: Restricted domain unavailable

Environment

  • Red Hat Enterprise Linux (RHEL) 5 or 6 with the High Availability Add On
    • NOTE: The error messages in the above examples cite "clurgmgrd", which is RHEL 5-specific. Similar messages or behavior may be seen in RHEL 6 coming from "rgmanager".
  • rgmanager
  • One or more services referencing a <failoverdomain/> that has <failoverdomainode/> elements
        <failoverdomain name="1then2" nofailback="1" ordered="1">
                <failoverdomainode name="node1.example.com" priority="1"/>
                <failoverdomainode name="node1.example.com" priority="2"/>
        </failoverdomain>

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.

Current Customers and Partners

Log in for full access

Log In

New to Red Hat?

Learn more about Red Hat subscriptions

Using a Red Hat product through a public cloud?

How to access this content