Re: [HACKERS] Tuple length limit - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: [HACKERS] Tuple length limit
Date
Msg-id 199907041420.KAA07003@candle.pha.pa.us
Whole thread Raw
In response to Re: [HACKERS] Tuple length limit  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
> Bruce Momjian <maillist@candle.pha.pa.us> writes:
> >> Can someone tell me what the maximum tuple length is?
> 
> I had always thought that the limit was supposed to be BLCKSZ less
> overhead.
> 
> > Here is what I found with the new code.  Seems it works.
> >     test=> vacuum;
> >     VACUUM
> 
> Wasn't the complaint that started this thread something about "peculiar
> behavior" of VACUUM with big tuples?  Might be wise to check VACUUM more
> closely.

We were inconsistent.  Varchar and vacuum where BLCKSZ/2, while others
where BLCKSZ, of course minus overhead.  The new code is consistent, and
does proper padding.  I even got rid of a fudge factor in rewrite
storage by using the actual rewrite lengths.

Will be in 6.5.1.

--  Bruce Momjian                        |  http://www.op.net/~candle maillist@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: Tom Lane
Date:
Subject: Re: [INTERFACES] Time and microseconds?
Next
From: Bruce Momjian
Date:
Subject: Re: Tuple too big