[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ocfs2-users
Subject:    Re: [Ocfs2-users] Run OCFS2 standalone with pacemaker
From:       Eric Ren <zren () suse ! com>
Date:       2016-07-09 2:09:48
Message-ID: 57805CEC.4010808 () suse ! com
[Download RAW message or body]

Hi Carlos,

On 07/08/2016 04:15 AM, Carlos Xavier wrote:
> Hi.
> 
> We had a cluster running fine for quite long time, One of the servers was lost \
> (burned) and I need to have access to the data on the remaining server.
> 
> OCFS2 was running over a DRBD device, I was able to load the module and start the \
> devices already. I could not get the DLM started yet. Is it possible to mount the \
> OCFS2 filesystems standalone?

No, unless you have "---fs-features=local" when mkfs, apparently which 
is not your case.

I'd like to help you get DLM back, and seemly it's the only way out.
1. How many nodes of your cluster? and how many remaining?
2. Please show your corosync configuration - /etc/corosync/corosync.conf
3. And your resource agents configuration - crm configure show

I guess you may have only one remaining node, which cannot form quorum. So,
dlm resource agent cannot be started? If yes, you can change 
corosync.conf to
have "two_node: 1" and "wait_for_all: 0" according to "man votequorum". 
And then
try to start pacemaker again, or try "corosync-cfgtool -R" when cluster 
is already running.
Hope this can help;-)

Eric

> Please, can somebody tell me how can I achieve this?
> 
> Best regards,
> Carlos
> 
> 
> 
> 
> _______________________________________________
> Ocfs2-users mailing list
> Ocfs2-users@oss.oracle.com
> https://oss.oracle.com/mailman/listinfo/ocfs2-users
> 


_______________________________________________
Ocfs2-users mailing list
Ocfs2-users@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-users


[prev in list] [next in list] [prev in thread] [next in thread]