We have just migrated to Postgres for a small deployable set of
applications.  The deployed system can be anywhere from 2 to 10+ nodes.

Our first cut at implementing replication is on a 2 node system (keeping it
simple) and this works fine.

We are two  implementing failure strategies using Slonik scripts:

 

1.       Soft change over master when all nodes are present.  This works
fine by using "move set"

2.       Hard change over master when the existing master node is not
present.  Trying to use "failover" but no go.

 

 In the soft case the commands used are:

 

lock set (id = 1, origin = 1);

wait for event (origin = 1, confirmed = 2);

move set (id = 1, old origin = 1, new origin = 2);

wait for event (origin = 1, confirmed = 2);

 

In the hard case the command is simply:

 

failover (id=1, backup node = 2)

 

The failover command does not seem to have any effect at all.

 

I have not been able to find the slon error log (Windows platform) to
determine why.

 

Any suggestions??

 

Thx Ted

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.slony.info/pipermail/slony1-general/attachments/20090115/d75ae613/attachment.htm