| From: | Scott Ribe <scott_ribe(at)elevated-dev(dot)com> |
|---|---|
| To: | "Graeme B(dot) Bell" <grb(at)skogoglandskap(dot)no> |
| Cc: | Postgres Maillist <pgsql-admin(at)postgresql(dot)org> |
| Subject: | Re: postgres hot-standby questions. |
| Date: | 2015-03-26 16:48:03 |
| Message-ID: | CD1B6BE2-8D28-4AE3-805F-7D42F4EA3ACA@elevated-dev.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-admin |
On Mar 26, 2015, at 10:40 AM, Graeme B. Bell <grb(at)skogoglandskap(dot)no> wrote:
>
> I could be missing something obvious, but how does this stop e.g. checkpoints, autovacuum etc from occuring and generating new wal that the standby might like? It seems like stopping services accessing the DB is likely to cause a lengthy autovacuum to begin, for example.
OK, fair point. But do you care?
I care that all committed transactions are on the replica. I don't care if they're still in WAL or written to the backing files. Actually, think about that question, you don't either--WAL gets copied to the replica, then moved to the backing files. Checkpoints don't affect replication as far as I know. And autovacuum? Couldn't care less if its effects are replicated yet; when the replica becomes master, autovacuum will run on it.
--
Scott Ribe
scott_ribe(at)elevated-dev(dot)com
http://www.elevated-dev.com/
https://www.linkedin.com/in/scottribe/
(303) 722-0567 voice
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Graeme B. Bell | 2015-03-26 18:17:50 | Re: postgres hot-standby questions. |
| Previous Message | Graeme B. Bell | 2015-03-26 16:40:19 | Re: postgres hot-standby questions. |