Sync cluster nodes configuration - lack of communication

Hi everybody,
I need help…

Two nodes cluster, nodeA and nodeB.
NodeB has been down for a week; during this week the cluster configuration on nodeA changed.
How can I synchronize nodeB with nodeA?

I tried with these commands but i did not solve.
On nodeA

crm configure show > configuration.txt

From the configuration.txt I deleted the option “last-lrm-refresh”
I copied configuration.txt on nodeB
On nodeB
#cibadmin -E --force
#crm configure < configuration.txt

The nodes are blind to each other.

Thanks
Mauro

On Fri, 21 Feb 2014 13:24:01 +0000, piscit61 wrote:
[color=blue]

Two nodes cluster, nodeA and nodeB.
NodeB has been down for a week; during this week the cluster
configuration on nodeA changed.[/color]

That seems to me to be a really bad idea.

[color=blue]

How can I synchronize nodeB with nodeA?[/color]

Bring nodeB back up?

[color=blue]

I tried with these commands but i did not solve. On nodeA

crm configure show > configuration.txt

From the configuration.txt I deleted the option “last-lrm-refresh” I
copied configuration.txt on nodeB
On nodeB
#cibadmin -E --force
#crm configure < configuration.txt[/color]

Why did you do this?

[color=blue]

The nodes are blind to each other.[/color]

Bring nodeB back up and fix the communications between them, should have
worked. Now, after ‘cibadmin -E --force’, I don’t know if that’s going to
be fixable without rebuilding the cluster. If you have the saved
configuration, it might be best to clean up both nodes, fix whatever
communications problem you have between them, then recreate the resources.

David Gersic dgersic_@_niu.edu
Knowledge Partner http://forums.netiq.com

Please post questions in the forums. No support provided via email.
If you find this post helpful, please click on the star below.

Hi David,
the configuration is saved.

How can I clean up both nodes?
I was thiking to shot “cibadmin -E --force” on both nodes and create from scratch /etc/corosync/corosync.conf. Is It right?

Thanks
Mauro

On Sat, 15 Mar 2014 07:44:01 +0000, piscit61 wrote:
[color=blue]

Hi David,
the configuration is saved.[/color]

Ok, that’s good to hear.

[color=blue]

How can I clean up both nodes?
I was thiking to shot “cibadmin -E --force” on both nodes and create
from scratch /etc/corosync/corosync.conf. Is It right?[/color]

I guess. I don’t know for sure, having not tried it. Fix the comms first.
I figure at this point, you’ve already broken it, so there’s not much for
you to lose. Give it a shot. If it doesn’t work, you’re no worse off than
you are now.

David Gersic dgersic_@_niu.edu
Knowledge Partner http://forums.netiq.com

Please post questions in the forums. No support provided via email.
If you find this post helpful, please click on the star below.