kevin kempter kevin at kevinkempterllc.com
Fri Sep 19 12:51:34 PDT 2008
Hi List;

I'm looking in the docs with little luck - could be that I'm just slow  
today.  I want to be able to do the following:


1) drop a node from a cluster. - I assume this is nothing more than  
executing a slonik 'drop node' ?

2) add a new node to an existing cluster - not sure how to do this

3) I have a failover script I'm testing. The failover script does a  
'failover' followed by a 'drop node'
depending on how the master comes down I end up with one of 2 states.

a) If I do a pg_ctl stop then the failover does the failover and the  
drop node, so when I restart the node the slony schema is removed.
      I assume in this case to re-add the node is the same process as  
adding a new node (#2 above), correct ?

b) If i do a pg_ctl -m immediate stop on the master then the failover  
does the failover but it seems that the drop node never executes  
because when I restart the cluster the slony schema is still there. Is  
there a different way to re-add this node to the cluster ?

Thanks in advance




More information about the Slony1-general mailing list