| From: | Chris Jewell <c(dot)jewell(at)lancaster(dot)ac(dot)uk> |
|---|---|
| To: | pgsql-admin(at)postgresql(dot)org |
| Subject: | WAL log archive frequency |
| Date: | 2006-01-19 10:44:35 |
| Message-ID: | 43CF6D93.2080806@lancaster.ac.uk |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-admin |
Hi,
Further to my question on PITR, I have now implemented it :-) However,
I was wondering about the frequency of archiving the WAL. Does
postgresql wait until the current WAL file has reached 16MB before
calling the archive_command? The reason why I ask is that last night,
after doing the base backup, I noticed that a WAL file had been written
to my backup server. This morning, I added a primary key to a table but
noticed that nothing was written to the backup server. Should it have
been?
Is there any method of forcing a WAL segment to be archived at a given
frequency irrespective of whether or not it is a full 16MB? Our
database changes so infrequently at the moment that I'm worried that
these small changes may not be backed up regularly by the PITR system.
Thanks,
Chris
--
Chris Jewell, BSc(Hons), BVSc, MRCVS
Dept of Maths and Statistics
Fylde College
Lancaster University
Lancaster
Lancs
LA1 4YF
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Juan Miguel | 2006-01-19 14:46:25 | Distributed database with postgres |
| Previous Message | Gourish Singbal | 2006-01-19 10:40:54 | Re: effects of cancelling a vacuum full |