-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Mon, Jun 09, 2003 at 02:33:30PM -0700, Jamin Roth wrote:
> OK running dmesg shows that the OS is out of memory.  When $sth->execute is
> run are all of the records returned?  That would take out a ton of memory.

It would not surprise me at all if MySQL was stupid enough to attempt this.

I initially figured it might be something in the script, but I looked over it
and I can't say I'd write it any differently.  It shouldn't be hanging on to
more than one row at a time.

It likely has nothing to do with file size, either.  The strace you sent shows
the process reading when it dies.  Does it manage to output anything before it 
gets terminated?  If not then that (unfortunately) supports the theory that the
MySQL client is attempting to suck down a ridiculously large number of records 
(maybe the entire result set) into local memory upon execute.

To the list: is it possible that the DBD is calculating a ludicrous initial size
for RowCacheSize?  Does MySQL even support a row cache?  That is my only guess
here.

A high-level DBI trace might confirm that.

- -- 
Stephen Clouse <[EMAIL PROTECTED]>
Senior Programmer/DBE, Core Technology Developer
The IQ Group, Inc. <http://www.theiqgroup.com/>

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE+5QEaA4aoazQ9p2cRAp2ZAKD0wyk0HfmZW0K6f9JpdPPuV0KSzgCg0HWa
SMSw4L0Oy1pExxP45QSpw04=
=QjBt
-----END PGP SIGNATURE-----

Reply via email to