Re: table level locking different in 7.0? - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: table level locking different in 7.0?
Date
Msg-id 200005180231.WAA23619@candle.pha.pa.us
Whole thread Raw
In response to Re: table level locking different in 7.0?  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
> > Jim Mercer <jim@reptiles.org> writes:
> >> i had several concurrent processes which would do inserts via COPY and
> >> queries.
> >> on that system, i don't recall the COPY processes as being blocked by the 
> >> query processes.
> >> now i'm running that app on solaris 7 with pgsql 7.0.
> >> i'm finding that a big long select is blocking other processes which
> >> are doing COPY's.
> 
> > Hmm.  In 7.0, COPY IN acquires an exclusive lock on the target table,
> > which is something I put in in a fit of paranoia.  It may not really
> > be necessary --- probably a regular write lock would be good enough.
> 
> OK, fix committed.  Jim, if you're in a hurry for this fix, just change
> AccessExclusiveLock to RowExclusiveLock at line 289 of
> backend/commands/copy.c.

FYI, I have been telling people to grab tomorrow's snapshot from
ftp:/pub/dev if they need changes that have been applied.  At this
point, we don't have any funny stuff in the cvs tree.

--  Bruce Momjian                        |  http://www.op.net/~candle pgman@candle.pha.pa.us               |  (610)
853-3000+  If your life is a hard drive,     |  830 Blythe Avenue +  Christ can be your backup.        |  Drexel Hill,
Pennsylvania19026
 


pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: table level locking different in 7.0?
Next
From: Tom Lane
Date:
Subject: Re: question about index cost estimates