Re: Reclaiming space for dropped database

From: Jerry Sievers <gsievers19(at)comcast(dot)net>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Alban Hertroys <alban(dot)hertroys(at)apollovredestein(dot)com>, pgsql-general(at)postgresql(dot)org
Subject: Re: Reclaiming space for dropped database
Date: 2019-01-23 19:36:56
Message-ID: 87o987upwn.fsf@jsievers.enova.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:

> Alban Hertroys <alban(dot)hertroys(at)apollovredestein(dot)com> writes:
>
>> Our current development database server is running a bit low on diskspace,
>> so I dropped an old but rather large database with the intention of
>> claiming back some space. However, the space remains claimed.
>> This server was upgraded from PG10 to PG11 using pg_upgrade's --link
>> option.
>
> If you used --link, then all the files would remain hard-linked from both
> the old and new database directories. You've got to remove them from the
> old DB directory as well.
>
> There's not really any point in keeping around the source DB directory
> once you've completed a --link migration. Starting the postmaster in
> the old DB directory would be disastrous because the files are
> inconsistent from its standpoint once the new postmaster has modified
> them at all. (In fact, I think pg_upgrade intentionally makes the old
> directory non-runnable to prevent that error.) So you might as well

Yeah. IIRC, it renames control to pg_control.old to avoid accidental
startup.

> just "rm -rf ./10", not only its biggest subdirectory.
>
> regards, tom lane
>
>

--
Jerry Sievers
Postgres DBA/Development Consulting
e: postgres(dot)consulting(at)comcast(dot)net

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Alex Morris 2019-01-24 02:08:22 how to properly start postgresql with no TCP listeners in ubuntu 16.04 LTS
Previous Message Alan Hodgson 2019-01-23 18:58:04 Re: Monitoring PITR recovery progress