Article ID: 123586, created on Nov 18, 2014, last review on Jan 14, 2016

  • Applies to:
  • Virtuozzo 6.0

Symptoms

After one of the PCS nodes was reinstalled and high availability was enabled again, "shaman stat" shows 2 resources for the node, one of them "Inactive":

# shaman stat
 Cluster 'mycluster'
 Nodes: 4
 Resources: 64

    NODE_IP           STATUS      RESOURCES
 *  10.20.3.11        Active      3 CT, 0 ISCSI, 5 VM
  M 10.20.3.12        Active      25 CT, 1 ISCSI, 14 VM
    10.20.3.11        Inactive    3 CT, 0 ISCSI, 1 VM
    10.20.3.13        Active      1 CT, 0 ISCSI, 2 VM

Cause

High Availability was not disabled on the node before it was reinstalled, and the old record remained.

Resolution

Find the "NODE_ID" for the inactive node using the "shaman -c <clustername> -v stat" command and then use "shaman -c <clustername> leave -N <node_id>" command to delete the node, e.g:

# shaman -c mycluster -v stat | head
 Cluster 'mycluster'
 Nodes: 4
 Resources: 64

    NODE_IP           STATUS      NODE_ID            RESOURCES
 *  10.20.3.11        Active      90d07219ebc648a4   3 CT, 0 ISCSI, 5 VM
  M 10.20.3.12        Active      00824984feef4fb3   25 CT, 1 ISCSI, 14 VM
    10.20.3.11        Inactive    96ad7a2abcea41e3   3 CT, 0 ISCSI, 1 VM
    10.20.3.13        Active      b6cc51e6aa1244f5   1 CT, 0 ISCSI, 2 VM

# shaman -c mycluster leave -N 96ad7a2abcea41e3

Search Words

clear old resource

inactive

shaman duplicate

clear missing resource

c62e8726973f80975db0531f1ed5c6a2 2897d76d56d2010f4e3a28f864d69223 0dd5b9380c7d4884d77587f3eb0fa8ef

Email subscription for changes to this article
Save as PDF