| From: | Andres Freund <andres(at)anarazel(dot)de> |
|---|---|
| To: | Simon Riggs <simon(at)2ndQuadrant(dot)com> |
| Cc: | Michael Paquier <michael(dot)paquier(at)gmail(dot)com>, Peter Eisentraut <peter_e(at)gmx(dot)net>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: pg_basebackup and replication slots |
| Date: | 2015-07-29 12:54:01 |
| Message-ID: | 20150729125401.GE24218@alap3.anarazel.de |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 2015-07-29 13:45:22 +0100, Simon Riggs wrote:
> So this would be needed when creating a standalone backup that would not be
> persistently connected to the master, yet we want to bring it up as a
> live/writable server in a single command
I'm not understanding what you mean with 'single command'
here. Currently there's no way to do this safely without configuring a
high enough wal_keep_segments.
You can (since yesterday) manually create a slot, run pg_basebackup, and
drop the slot. But that's not safe if your script is interrupted
somehow. Since many base backups are run in a non-interactive fashion
asking for intervention to clean up in that case imo is not an
acceptable answer.
> and we want to make it easy to script in case our script is killed?
Or the connection dies/is killed, or the server is restarted, or ...
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Stephen Frost | 2015-07-29 13:09:27 | Re: A little RLS oversight? |
| Previous Message | Simon Riggs | 2015-07-29 12:53:31 | Re: Don'st start streaming after creating a slot in pg_receivexlog |