[Veritas-ha] System Panic

[Gene Henriksen]

Check your /etc/llttab and /etc/llthosts.

The set-node should be unique within the 3 clusters.
Check the /etc/llthosts to make sure the 3 nodes are defined correctly.
The set-cluster line of /etc/llttab should have a unique cluster ID for each
cluster.

Panic is a method of preventing split brain when multiple nodes come up as
the same cluster ID and same node ID.

set-cluster 1
set-node 1
or set-node sysa    (this will point to a line in llthosts with sysa)

There is one other possibility, but it has to be set manually
set-node /etc/VRTSvcs/conf/sysname  (where the sysname file contains the
name to lookup in               the /etc/llthosts). 

> ----------
> From: 	Sandip Samanta
> Sent: 	Friday, February 7, 2003 6:14 AM
> To: 	veritas-ha at mailman.eng.auburn.edu
> Subject: 	[Veritas-ha] System Panic
> 
> <<File: ssamanta.vcf>>
> Hi,
> 
> We are facing some problem in VCS. We have configured 3 cluster in 9
> nodes(HP-UX 11i).
> Every cluster configured in 3 nodes.
> 
> But all 9 cluster keep on rebooting & showing the below messages..
> 
> Reboot after panic: GAB: Port a halting system due to node id conflict.
> 
> What could be reason.
> 
> Regds
> Sandip
> 
>