Re: integrate pg_upgrade analyze_new_cluster.sh into vacuumdb

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Peter Eisentraut <peter_e(at)gmx(dot)net>
Cc: pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: integrate pg_upgrade analyze_new_cluster.sh into vacuumdb
Date: 2014-01-09 03:32:26
Message-ID: 7737.1389238346@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> pg_upgrade creates a script analyze_new_cluster.{sh|bat} that runs
> vacuumdb --analyze-only in three stages with different statistics target
> settings to get a fresh cluster analyzed faster. I think this behavior
> is also useful for clusters or databases freshly created by pg_restore
> or any other loading mechanism, so it's suboptimal to have this
> constrained to pg_upgrade.

> Therefore, I suggest to add this functionality into the vacuumdb
> program.

Seems reasonable.

> There are some details to be considered about who pg_upgrade would call
> this. For example, would we keep creating the script and just have the
> script call vacuumdb with the new option, or would we skip the script
> altogether and just print a message from pg_upgrade? Also, pg_upgrade
> contains logic to run a vacuum (not only analyze) in the final run when
> upgrading from PostgreSQL <8.4 to deal with the freespace map. Not sure
> how to adapt that; maybe just keep the script and run a non-analyze
> vacuum after the analyze.

I don't think this vacuumdb feature should deal with any
version-conversion issues. So it sounds like the thing to do is keep the
wrapper script, which will give us a place to put any such special actions
without having to kluge up vacuumdb's behavior. That'll avoid breaking
scripts that users might've built for using pg_upgrade, too.

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Mark Dilger 2014-01-09 03:34:25 Re: In-core regression tests for replication, cascading, archiving, PITR, etc. Michael Paquier
Previous Message Alvaro Herrera 2014-01-09 03:27:21 Re: Add CREATE support to event triggers