[Pacemaker] Several problems with simple virtual-IP config

Klaus Darilion klaus.mailinglists at pernau.at
Wed Feb 9 08:16:23 UTC 2011


>> 3. Now, server1, hosting the virtual-IP, loost connectivity to the ping
>> target (I inserted a firewall rule) -> The virtual-IP stayed with server1.
>>
>> Now I put server2 online again: # crm node online server2.
>> That means, server2 is online and has ping connectivity, server1 is
>> online and doesn't have ping connectivity. But the virtual-IP stayed
>> with server1:
>>
>>   Online: [ server1 server2 ]
>>
>>   failover-ip     (ocf::heartbeat:IPaddr):        Started server1
>>    Clone Set: clonePing
>>        Started: [ server2 server1 ]
>>
>> What do I have to change in the config to have here a failover to server2?
> 
> What are the pingd scores for each node? You can display it with
> crm_mon -1f

Hi Michael!

Following is the same scenario (same config) again, this time with ping
scores:

1. Corosync on both servers restarted:

	============
	Last updated: Wed Feb  9 09:05:59 2011
	Stack: openais
	Current DC: server1 - partition with quorum
	Version: 1.0.9-74392a28b7f31d7ddc86689598bd23114f58978b
	2 Nodes configured, 2 expected votes
	2 Resources configured.
	============
	Online: [ server1 server2 ]
	
	failover-ip     (ocf::heartbeat:IPaddr):        Started server1
	 Clone Set: clonePing
	     Started: [ server2 server1 ]
	
	Migration summary:
	* Node server1:
	* Node server2:

Why there is no summery at all?

2. # crm node standby server2
(only server1 is pinging)

	Node server2: standby
	Online: [ server1 ]
	
	failover-ip     (ocf::heartbeat:IPaddr):        Started server1
	 Clone Set: clonePing
	     Started: [ server1 ]
	     Stopped: [ pingtest:0 ]
	
	Migration summary:
	* Node server1:  pingd=10
	* Node server2:  pingd=0

I think this looks fine.

3. # crm node online server2
(both servers are pinging again)

	Online: [ server1 server2 ]
	
	failover-ip     (ocf::heartbeat:IPaddr):        Started server1
	 Clone Set: clonePing
	     Started: [ server2 server1 ]
	
	Migration summary:
	* Node server1:  pingd=10
	* Node server2:  pingd=0

Why is pingd=0 for server2? I would expect it to be 10 too.

4. # crm node standby server2
(only server1 is pinging)

	Node server2: standby
	Online: [ server1 ]
	
	failover-ip     (ocf::heartbeat:IPaddr):        Started server1
	 Clone Set: clonePing
	     Started: [ server1 ]
	     Stopped: [ pingtest:0 ]
	
	Migration summary:
	* Node server1:  pingd=10
	* Node server2:  pingd=0

5. server1 pings will be blocked in the firewall, after 10 seconds:

	Node server2: standby
	Online: [ server1 ]
	
	failover-ip     (ocf::heartbeat:IPaddr):        Started server1
	 Clone Set: clonePing
	     Started: [ server1 ]
	     Stopped: [ pingtest:0 ]
	
	Migration summary:
	* Node server1:  pingd=0
	* Node server2:  pingd=0

Seems correct, as server1's pings are blocked and server2 is in standby
mode.

6. # crm node online server2
(both servers are pinging again, server1's pings are still blocked)

	Online: [ server1 server2 ]
	
	failover-ip     (ocf::heartbeat:IPaddr):        Started server1
	 Clone Set: clonePing
	     Started: [ server2 server1 ]
	
	Migration summary:
	* Node server1:  pingd=0
	* Node server2:  pingd=0

Here, it should switch to server2. And why is server2's pingd=0 instead
of 10?

7. # crm node standby server1
(only server2 is pinging)

	Node server1: standby
	Online: [ server2 ]
	
	failover-ip     (ocf::heartbeat:IPaddr):        Started server2
	 Clone Set: clonePing
	     Started: [ server2 ]
	     Stopped: [ pingtest:1 ]
	
	Migration summary:
	* Node server1:  pingd=0
	* Node server2:  pingd=0

Even if server1 is in standby, the pingd value of server2 is 0 instead
of 10.

What am I doing wrong?

Thanks
Klaus




More information about the Pacemaker mailing list