| From: | "Gregor Zeitlinger" <gregor(dot)zeitlinger(at)torexretail(dot)de> |
|---|---|
| To: | "Zach Bagnall" <zach(dot)bagnall(at)bulletinwireless(dot)com>, <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Incremental Backup Script |
| Date: | 2006-01-09 11:03:46 |
| Message-ID: | 5DE489C997EC984FA3DD0935879DAE1255EB26@ex09-00-z002.torexretail.de |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
> As per docs, if the databases are rarely updated it could take a long
> time for the WAL segment to "roll over".
Yes, therefore I want to copy the current WAL (as I said earlier).
When restoring, I also want to make sure that I restore exactely to the point when I copied the current WA segment.
Hence I consider to do it as follows:
1) take the t = current time
2) copy the current WAL
3) when restoring, set recovery_target_time = t
Maybe there is even a way to ask Postgres of its last commited x = xid.
In that case, we could set recovery_target_xid = x
Is that possible?
Regards,
Gregor
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Michael Glaesemann | 2006-01-09 11:45:40 | Re: ISO 8601 Intervals |
| Previous Message | Lukas Smith | 2006-01-09 10:52:16 | Re: Improving N-Distinct estimation by ANALYZE |