Date:         Mon, 4 Nov 1996 15:07:12 -0800
Reply-To:     Paul OldenKamp <oldenkmp@REDWOOD.RT.CS.BOEING.COM>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Paul OldenKamp <oldenkmp@REDWOOD.RT.CS.BOEING.COM>
Organization: Boeing Commercial Airplane Group
Subject:      Re: Crontab & SAS Batch Submits
Comments: To: Roger Bolton <rkbolt@NOVA.MONSANTO.COM>

Roger Bolton wrote:
>
> I run quit a bit of SCL code with CRON.  However, for some reason, in
> some jobs it can get to a certain part of my SAS job and just quit.
> I"ve juet to track down why.
 
One thing that I would look at is the use of memory.  SCL programs
can take a lot of memory and if you run out of swap space it can appear
like the SAS program has just quit since the last part of the log is
still in the buffer (try the unbuflog(sp?) option to get all of it in the
log file).
 
I have just pinned down that large simulation jobs were failing here
because of a memory leak in the 6.11 data step on Solaris.  They
appeared to just stop when all the swap space (270 MB!!!) was used
up.
 
Paul OldenKamp, Boeing, Seattle, WA