Re: [BUGS] Bug #655: win32 client and bytea column - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: [BUGS] Bug #655: win32 client and bytea column
Date
Msg-id 200206022153.g52LrZ908624@candle.pha.pa.us
Whole thread Raw
List pgsql-hackers
Krzysztof Stachlewski wrote:
> ----- Original Message -----
> From: "Bruce Momjian" <pgman@candle.pha.pa.us>
> To: <stach@toya.net.pl>; <pgsql-bugs@postgresql.org>
> Sent: Sunday, June 02, 2002 10:45 PM
> Subject: Re: [BUGS] Bug #655: win32 client and bytea column
> 
> > Yes, this is certainly our error message:
> >
> > pg_exec() query failed: pqReadData() -- read() failed: errno=0 No error
> >
> > Of course, the "0 No error" is odd.  I think that just means that read()
> > itself didn't set errno for the failure.
> >
> > You are clearly base64 encoding the storage info, meaning it isn't some
> > strange character in the data.  Are you using 7.2 for the server and the
> > client?
> 
> Yes. libpq.dll is version 7.2.1. So is the server.
> 
> > My guess is that Win isn't handling some of the larger packets,
> > but I may be wrong.  If it fails reliably, could you find the exact
> > length where it fails.  That may help.
> 
> It is exactly 6106 bytes of raw data.
> That is 8144 bytes of base64 encoded data.
> Below this everything works just fine.

Could it be that Win32 somehow can't handle large tcp packets?  Would
you create a TEXT field, and stuff 9k of a string into it and see if
that fails?

--  Bruce Momjian                        |  http://candle.pha.pa.us 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: perl pod2man bug
Next
From: Bruce Momjian
Date:
Subject: Re: Trying to reduce per tuple overhead (bitmap)