From owner-freebsd-questions  Sun May  4 11:54:56 1997
Return-Path: <owner-questions>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id LAA10326
          for questions-outgoing; Sun, 4 May 1997 11:54:56 -0700 (PDT)
Received: from iceberg.anchorage.net. (root@iceberg.anchorage.net [207.14.72.150])
          by hub.freebsd.org (8.8.5/8.8.5) with SMTP id LAA10317
          for <freebsd-questions@FreeBSD.ORG>; Sun, 4 May 1997 11:54:49 -0700 (PDT)
Received: from aak.anchorage.net (ai-129 [207.14.72.129]) by iceberg.anchorage.net. (8.6.11/8.7.3) with SMTP id JAA08945; Sun, 4 May 1997 09:51:06 -0800
Date: Sun, 4 May 1997 01:48:08 -0800 (AKDT)
From: Steve Howe <un_x@anchorage.net>
X-Sender: abc@aak.anchorage.net
To: Steve <shovey@buffnet.net>
cc: freebsd-questions@FreeBSD.ORG
Subject: Re: Weird perl problem
In-Reply-To: <Pine.BSI.3.95.970504102416.23735T-100000@buffnet11.buffnet.net>
Message-ID: <Pine.BSF.3.95q.970504014424.21505B-100000@aak.anchorage.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-questions@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

On Sun, 4 May 1997, Steve wrote:

i noticed while opening a few Megs of text files with "joe" editor,
my terminal went completely nut, shell and everything, and i have
to exit to the login prompt to regain control.  i'm not an expert
on BSD RAM usage, but i something goes nuts when it gets amxed out.
i think perl sucks in that file as one long string into memory.
i wonder if it would crash if you chopped the file in 2 or 4 ...

> On both 2.1R and 2.1.5R boxes, Im having this really really weird problem
> with perl.
> 
> I run wwwstat against apache logs.  Lately on a main large install (the
> logs being close to or ovre 400MB) the perl script will run, print out
> 8192 bytes exactly of report output, then abort with a floating point
> exception error message.
> 
> I thought it might be some kind of divide error and wasted weeks putting
> in checks before any math operations.
> 
> Then finally I put in an echo check - just a line to spew out some
> iteration information, and the point the program stopped at changed, I
> mean what secton of the report it was on when it aborted - but again, the
> report itself was exactly 8192 bytes in size.
> 
> The report goes to stdout which I use >filename to trap in a file.
> 
> Any ideas?  Im going crasy!
> 
> 
> 

-------------------------------------------------------------------------
 Sleep: a sign a caffeine deprivation ... http://www.anchorage.net/~un_x
-------------------------------------------------------------------------