[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

RE: [Linux-cluster] Can't create new GFS storage



You might check that /etc/lvm/lvm.conf has locking_type = 3 as described in FAQ #22 here.  It defaults to local locking.
http://sources.redhat.com/cluster/faq.html

Also, I've found that cluster suite components use a variety of IP ports for communication, and if you've selected a standard firewall configuration you may be blocking them.  For testing, I'd try shutting down the CS services (service cman stop, etc), then run "iptables -F" on all nodes to flush the rulesets, then restart the CS services on all nodes.  You can discover what ports are being used by cluster suite by doing a netstat -na before and after this process and compare the ports that are in use, then put together a custom /etc/sysconfig/iptables file if you want.

Also, assuming you're using a Redhat-based distro (Fedora, CentOS, RH), you can check the status of the CS services with the following rather than using "ps":
# service cman status
# service clvmd status
etc.

One other tool to check your cluster's status is cman_tool.  Do a man on it if you haven't used it.

HTH
Doug


-----Original Message-----
From: linux-cluster-bounces redhat com on behalf of isplist logicore net
Sent: Mon 1/28/2008 6:35 PM
To: linux-cluster
Subject: RE: [Linux-cluster] Can't create new GFS storage
 
> 2. # pvcreate /dev/sdb

This is happening from any node now. If I try to pvcreate, it just sits there.
I am seeing in the messages log;

Jan 28 15:42:52 img62 rgmanager: clurgmgrd startup failed

Anyone have any thoughts on what might be causing this? I don't seem to have 
much for errors and google doesn't seem to help much.

Mike



--
Linux-cluster mailing list
Linux-cluster redhat com
https://www.redhat.com/mailman/listinfo/linux-cluster

<<winmail.dat>>


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]