[Ocfs2-users] Version mis-match?

[Sunil Mushran]

If you can ping the nodes on the private interface (ping -I ethX node)
then see if iptables is running. If so, stop it and attempt mount again.

Also, 1.2.3 and 1.2.5 are not compatible. Upgrade 1.2.3 to 1.2.5.
Best would be to upgrade them all to 1.2.8.

Kendall, Kim wrote:
> Dmesg error:
> ------------
> (3646,1):o2net_connect_expired:1563 ERROR: no connection established
> with node 3 after 10.0 seconds, giving up and returning errors.
> (10386,1):dlm_request_join:902 ERROR: status = -107
> (10386,1):dlm_try_to_join_domain:1050 ERROR: status = -107
> (10386,1):dlm_join_domain:1322 ERROR: status = -107
> (10386,1):dlm_register_domain:1515 ERROR: status = -107
> (10386,1):ocfs2_dlm_init:2011 ERROR: status = -107
> (10386,1):ocfs2_mount_volume:1083 ERROR: status = -107
> ocfs2: Unmounting device (120,32) on (node 2)
>
> The existing 3 nodes are Oracle Enterprise Linux 4 Update 5
> The ocfs packages are what is part of the distro:
> #rpm -qa | grep ocfs
> ocfs2-tools-1.2.4-1
> ocfs2-2.6.9-55.0.0.0.2.ELsmp-1.2.5-1
> ocfs2console-1.2.4-1
>
> Same with the 4th node - OEL4 U4
> #rpm -qa | grep ocfs
> ocfs2console-1.2.2-2
> ocfs2-tools-1.2.2-2
> ocfs2-2.6.9-42.0.0.0.1.ELlargesmp-1.2.3-2
>
> The errors seem to point to network, but I can verify connectivity on
> both public and private LANs.
>
> -Kim
>
>
> -----Original Message-----
> From: Joel Becker [mailto:Joel.Becker at oracle.com] 
> Sent: Thursday, May 01, 2008 5:53 PM
> To: Kendall, Kim
> Cc: ocfs2-users at oss.oracle.com
> Subject: Re: [Ocfs2-users] Version mis-match?
>
> On Thu, May 01, 2008 at 05:41:05PM -0700, Kendall, Kim wrote:
>   
>> When I try to bring up any of the volumes on the other 3 nodes, I get:
>>     
>
>   
>> mount.ocfs2: Transport endpoint is not connected while mounting
>> /dev/emcpowerc on /oradata/prod/datafiles/1.
>>
>> The 3 existing systems are Oracle Enterprise Linux 4 Update 5
>> The new system is Oracle Enterprise Linux 4 Update 4
>>
>> Could the ocfs version mis-match be the cause? The error looks like
>>     
> the
>   
>> private network interconnect is the problem.
>>     
>
> 	Well, what version of ocfs2 are you running?  Specifically, the
> output of:
>
> # uname -a
> # rpm -qa | grep ocfs2
>
> If you can give us that detail for all four nodes, we can hazard a
> guess.  You also might want to check dmesg(8) to see if there is any
> details for the error.
>
> Joel
>
>