[Linux-cluster] How to disable node?

[Rick Stevens]

Rick Stevens wrote:
> Jakov Sosic wrote:
>> On Mon, 31 Aug 2009 15:30:13 -0500
>> Alan A <alan.zg at gmail.com> wrote:
>>
>>> You have to give your cluster 2 node setting on remaining two nodes.
>>>
>>> How can I define a two-node cluster if a majority is needed to reach
>>> quorum?
>>>
>>> We had to allow two-node clusters, so we made a special exception to
>>> the quorum rules. There is a special setting "two_node" in
>>> the /etc/cluster.conf file that looks like this:
>>>
>>> <cman expected_votes="1" two_node="1"/>
>>>
>>> This will allow one node to be considered enough to establish a
>>> quorum. Note that if you configure a quorum disk/partition, you don't
>>> want two_node="1".
>>>
>>> http://sources.redhat.com/cluster/wiki/FAQ/CMAN#two_node
>>
>> Yeah I know how to set up two_node cluster, but didn't know that
>> reconfiguration is the only way out when one node is missing :( I'm
>> disappointed with this limitation, I hope that developers will sort it
>> out for RHEL6 :(
> 
> I don't see that there's anything to fix.  You had a three-node cluster
> so you needed a majority of nodes up to maintain a quorum.  One node
> died, killing quorum and thus stopping the cluster (the expected and
> correct behavior).  As a three-node cluster, it's dead.  It can't be run 
> as a three-node cluster until the third node is fixed.  Those are the
> rules.
> 
> A two node cluster requires special handling of things to prevent the
> dread split-brain situation, which is what two_node does.  Running the
> surviving nodes as a two-node cluster is, by definition, a
> reconfiguration.  I'd say simply requiring you to set two_node is pretty
> damned innocuous to let you run a dead (ok, mortally wounded) cluster.
> 
> If you pulled a drive out of a RAID6--thus degrading it to a RAID5--
> would you complain because it didn't remain a RAID6?

On re-reading my response, it seemed unintentionally harsh.  I didn't
mean any disrespect, sir.  I was simply questioning the concept that a
reconfiguration of a cluster shouldn't be required when, indeed the
cluster was being reconfigured.  The other response I saw to this
thread regarding planning, and things such as last-man-standing was much
better worded.

My apologies if it seemed I was jumping down your throat.  I wasn't.
----------------------------------------------------------------------
- Rick Stevens, Systems Engineer                      ricks at nerd.com -
- AIM/Skype: therps2        ICQ: 22643734            Yahoo: origrps2 -
-                                                                    -
-     Is that a buffer overflow or are you just happy to see me?     -
----------------------------------------------------------------------