Slony1 PostgreSQL Replication - maximum sync

You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

Author

maximum sync_group

Michael Crozier

2006-02-02, 3:25 am

Hi,

I'm curious why the maximum size of the sync_group array in remote_worker.c is
only 100 events. It seems odd and unpractical that slon doesn't obey the
"-g" option. I had to recompile slon today with an increased size in order
to get slony to "catch up" after event processing was halted due to my
"post-merge re-replication" problem.

I increased the size of the array and the corresponding if statement
conditions to 2000. While this is obviously a very large amount of data, the
slon daemons didn't inflate due to the incremental fetches from the cursor.
Instead, this reduced the number of index/table scans over sl_log_1 by 20x
with only a 2x time penalty. Slony changed from falling behind at a dramatic
rate to catching up on 3 million transactions in only a few hours.

Regards,
Michael

Sponsored Links