Created attachment 609605 [details] /etc/cluster/cluster.conf Description of problem: When there is delay=15 added into <device> node in cluster.conf it is ignored by /sbin/fence_node. fence_node command should first wait $delay seconds before trying to send fence_command which it does not at the moment. Version-Release number of selected component (if applicable): RHEL-5.9, cman-2.0.115-106.el5 How reproducible: always. Steps to Reproduce: 1. have cluster of three virtual machinech 2. on one of the nodes create/modify /etc/cluster/cluster.conf similar to the one attached in which each <device> will have delay=15 attribute. 3. use "ccs_tool update /etc/cluster/cluster.conf" to distribute this config to all nodes on the cluster. 4. on one of the nodes time fencing other node in the cluster Actual results: takes about two seconds. Expected results: should first wait $delay seconds and then try to fence the node which should take more than 15 seconds. Additional info: partly related: https://bugzilla.redhat.com/show_bug.cgi?id=804170
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2013-1304.html