Re: Tricky bugs in concurrent index build

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Gregory Stark <stark(at)enterprisedb(dot)com>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Tricky bugs in concurrent index build
Date: 2006-08-27 19:18:14
Message-ID: 9645.1156706294@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

I wrote:
> I'm toying with the idea of adding a lock manager call defined as
> "give me a list of XIDs that currently hold locks conflicting with
> lockmode X on object Y" --- but not including XIDs merely waiting
> for such a lock. Then we could get the list of XIDs currently blocking
> ExclusiveLock, and do XactLockTableWait on each one.

I've committed a patch along these lines. I also reduced the strength
of the lock we check for from ExclusiveLock to ShareLock, which seems
sufficient --- did you have a reason for selecting ExclusiveLock in the
original coding?

regards, tom lane

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2006-08-27 19:44:16 Re: integration of pgcluster into postgresql
Previous Message Jonah H. Harris 2006-08-27 17:45:00 Re: integration of pgcluster into postgresql