[Linux-HA] Weird problem with HA and pingd

[Paulo F. Andrade]

Hi,

I'm trying to setup some ping nodes, but I'm currently running into  
some weird problems.
So i've setup 3 ping nodes, with a multiplier set to 1000.
- score for lvs1 is 1000
- score for lvs2 is 500
- resource_stickiness 100
- resource_failure_stickiness -500
(ha.cf and cib.xml are attached)

I start heartbeat and everything is fine, my resource is running on  
lvs1 and lvs2 is on standby.
Executing command "crm_attribute -V -G -t status -U lvs1 -n pingd"  
returns "name=pingd value=3000".

When I block the ping-reply from one of the servers to lvs1, the  
previous command returns "name=pingd value=2000"
as expected. But lvs1 goes offline and the resource starts on lvs2...  
even worse the resource isn't stopped on lvs1!!
I was expecting resources to migrate and lvs1 to still be online...

When I re-enable the ping-reply from the server, pingd value updates  
correctly but lvs1 continues offline.

If then I do a "crm_resource -C" the GUI says the resources are   
multi-running on both nodes!

Please help, as I am very confused about what's going on..
PS: Is there a way to see the scores of each resource on some instance?

Paulo F. Andrade 52439 at IST
mailto: pfca at mega.ist.utl.pt

-------------- next part --------------
A non-text attachment was scrubbed...
Name: cib.xml
Type: text/xml
Size: 5343 bytes
Desc: not available
Url : http://lists.community.tummy.com/pipermail/linux-ha/attachments/20061013/8494c6ee/cib.bin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ha.cf
Type: application/octet-stream
Size: 1517 bytes
Desc: not available
Url : http://lists.community.tummy.com/pipermail/linux-ha/attachments/20061013/8494c6ee/ha.obj
-------------- next part --------------