David Parker wrote:

> In the process of testing various failover/failback scenarios, I ran 
> into this error:
>  
> CLUSTER NAME = tzreplic;
>  
> NODE 2 ADMIN CONNINFO = 'host=192.168.111.65 port=24005 dbname=policy 
> user=tazz:dbmgr password=tazz:dbmgr';
> NODE 3 ADMIN CONNINFO = 'host=192.168.111.62 port=24005 dbname=policy 
> user=tazz:dbmgr password=tazz:dbmgr';
> LOCK SET (ID = 1, ORIGIN = 2);WAIT FOR EVENT (ORIGIN = 2, CONFIRMED = 
> 3);MOVE SET (ID = 1, OLD ORIGIN = 2, NEW ORIGIN = 3);WAIT FOR EVENT 
> (ORIGIN = 2, CONFIRMED = 3);
>  
> stdout:
>  
> <stdin>:5: Error: No admin conninfo provided for node 1<stdin>:5: 
> Error: No admin conninfo provided for node 1
> stderr:
>  
> This is with slony 1.0.2. After a failover when our code needs to 
> re-establish a cluster, it assigns the node id by simply incrementing 
> the max existing node id. So in the scenario above, we had failed over 
> once, dropping node 1, and then added the failed node back in as a new 
> standby with id = 3.
>  
> I seem to remember something about having to have a node 1, but it 
> doesn't make sense to me. Does the error indicate that I have some 
> misconfiguration elsewhere?
>  
> Thanks.

Try adding

wait on = 2

to the wait for event command.  Docs say that the default for this value 
is 1.

-- 
Brad Nicholson  416-673-4106    bnichols at ca.afilias.info
Database Administrator, Afilias Canada Corp.