Are you doing the lock set and wait for event slonik commands within the
same try block?

eg.

try {
LOCK SET (ID = 1, ORIGIN = 1);
WAIT FOR EVENT (ORIGIN = ALL, CONFIRMED = ALL, WAIT ON = 1);
}

This won't work, because the lock set and wait for event slonik commands
are part of the same transaction. It'll sit and wait forever.

If this isn't the problem, could you post your slonik scripts for us to
take a look at?

Tim

On Wed, Feb 16, 2005 at 11:12:44AM -0500, Fiel Cabral wrote:
> Initially: 
> Node 1 is the master. Node 2 is the slave.
> There is only one replication set (set ID 1).
> 
> Switchover is defined as:
> LOCK SET 1.
> WAIT FOR EVENT.
> MOVE SET 1 from node 1 to node 2.
> WAIT FOR EVENT.
> 
> I run the switchover script on node 2 to effect a switchover.
> After waiting for 1 hour I realize that switchover is stalled. I find
> that if I stop and then start the database clients that are connected
> to node 1, the switchover may complete successfully.
> 
> What are the factors that cause switchover to stall? What are the
> factors that help switchover to complete?
> 
> Thank you for any guidance in advance.
> -Fiel Cabral
> _______________________________________________
> Slony1-general mailing list
> Slony1-general at gborg.postgresql.org
> http://gborg.postgresql.org/mailman/listinfo/slony1-general

-- 
Tim Goodaire    416-673-4126    tgoodair at ca.afilias.info
Database Administrator, Afilias Canada Corp.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://gborg.postgresql.org/pipermail/slony1-general/attachments/20050217/20e1a3bf/attachment.bin