[Bucardo-general] Swap replication latency

[Greg Sabino Mullane]

> I've tried, but none of the operations I could infer were happening
> from the log seem to profile as taking more than a trivial amount of
> time on the main or secondary DB.

I meant the Bucardo log. There should be a KID process that is created 
or kicked, then gathers the deltas, makes the changes on both sides, 
and does a final commit. This is where the 40+ minutes is. You could also 
look at the Postgres logs, but things won't be as easy. But you could 
scan for SET TRANSACTION lines from the bucardo user, then trace it 
down until you hit the final commit. In addition to the time between 
statements is the time of each query, assuming you are logging durations. 
If you are not, I think it can be enabled in postgresql.conf without 
a restart, just a HUP.

-- 
Greg Sabino Mullane greg at endpoint.com
End Point Corporation
PGP Key: 0x14964AC8
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 163 bytes
Desc: not available
Url : https://mail.endcrypt.com/pipermail/bucardo-general/attachments/20110502/58e8ffde/attachment.bin