Re: [Linux-cluster] pull plug on node, service never relocates
From: Dustin Henry Offutt <dhoffutt gmail com>
To: linux clustering <linux-cluster redhat com>
Subject: Re: [Linux-cluster] pull plug on node, service never relocates
Date: Fri, 21 May 2010 17:14:58 -0500
Did I mention?
Regarding the state after pulling the plug...:
Watching clustat from another member of the cluster, the state of service core1 remained "started" until after node1 was fenced, the OS had rebooted, the node started its cluster software, and rejoined the cluster - at that time the state changed to "recoverable." Approximately 30 seconds after that the spare node started the service core1, during which time it was obviously "starting."
Going to try the latest cman (2.0.115-34) and rgmanager (2.0.52-6) from the RHEL5.5 stream now (am using RHEL5.4).
Just tried it. No luck. Installed a few more updates of various related things that get installed when doing a yum group-install clustering... Again, no change.
BUT just noticed something interesting - This is a five node cluster with four configured cluster services. If I only have three of the cluster services running, leaving two nodes "spare," then simulate a node failure, the node gets fenced AND the service immediatley relocates.....
Dirección Comercial: C/Jose Bardasano Baos, 9, Edif. Gorbea 3, planta
3ºD, 28016 Madrid, Spain
Dirección Registrada: Red Hat S.L., C/ Velazquez 63, Madrid 28001, Spain
Inscrita en el Reg. Mercantil de Madrid – C.I.F. B82657941