| From: | Stephen Frost <sfrost(at)snowman(dot)net> |
|---|---|
| To: | Peter Eisentraut <peter(dot)eisentraut(at)2ndquadrant(dot)com> |
| Cc: | Devrim Gündüz <devrim(at)gunduz(dot)org>, pgsql-pkg-yum <pgsql-pkg-yum(at)postgresql(dot)org> |
| Subject: | Re: Turning on archive_mode by default |
| Date: | 2020-05-20 16:59:26 |
| Message-ID: | 20200520165926.GI3418@tamriel.snowman.net |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-pkg-yum |
Greetings,
* Peter Eisentraut (peter(dot)eisentraut(at)2ndquadrant(dot)com) wrote:
> On 2020-05-20 16:21, Devrim Gündüz wrote:
> >Does anyone know whether there is an overhead of turning on
> >archive_mode, and setting archive_command to /bin/true?
>
> The overhead is probably small, but what this would do is start the archiver
> and report to stats views etc. that archiving is running and progressing,
> even though it's doing nothing. That seems pretty bogus and confusing.
I tend to agree with it being confusing. Simpler might be to just
depend on pgbackrest and automatically set up archiving and backups.
> Most users[citation needed] don't even use archiving anymore, so this is the
> wrong end of history to be fiddling with this setting.
No. Lots of users use archiving and until we've got a real answer to
being able to perform bulk archiving at scale, that's not likely to
change. pg_receivewal is absolutely not reasonable as a solution to
WAL archiving and management.
We've considered adding WAL streaming support to pgbackrest but it's not
much of a priority because it's rather ugly and not particularly better
than archive_command for most use-cases. Of course, we'd certainly
encourage folks to work with us to develop it and send us patches for
it.
Thanks,
Stephen
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Peter Eisentraut | 2020-05-20 22:31:49 | Re: Removing PL/Python2 from PostgreSQL 13 |
| Previous Message | Stephen Frost | 2020-05-20 16:35:09 | Re: Can we stop defaulting to 'ident'? |