On 24/02/2015 10:28, Clement Thomas wrote:
> Hi,
>     Thanks for pointing it. Ya the statement_timeout was set to 0 in
> node1 and 600000 in node2 and node3. when i set it to 0 on node2 and
> node3 replication lagged. and i could see
> 
> NOTICE:  Slony-I: log switch to sl_log_2 still in progress - sl_log_1
> not truncated
> 
> i had to set statement_timeout back to 600000 to fix the replication
> issue as the nodes are in production. what is the better way to clean
> these tables?

> 

AFAICT, your replication falls behind because it has a lot of work to do.
Setting a non-zero statement_timeout prevents it to fully propagate data changes.

-- 
Stéphane Schildknecht
Contact régional PostgreSQL pour l'Europe francophone
Loxodata - Conseil, expertise et formations
06.17.11.37.42

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: OpenPGP digital signature
Url : http://lists.slony.info/pipermail/slony1-general/attachments/20150318/1cdd46bb/attachment.pgp