Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!tiamat!jim
From: jim@tiamat.fsc.com ( IT Manager)
Newsgroups: comp.unix.sysv386
Subject: Re: SCO NFS dies when heavily used
Message-ID: <854@tiamat.fsc.com>
Date: 7 Jun 91 13:26:33 GMT
References: <1991Jun06.171047.15327@nss1.com>
Organization: Ahlstrom Filtration - Chattanooga,TN
Lines: 16

In article <1991Jun06.171047.15327@nss1.com>, mrm@nss1.simpact.com (Michael R. Miller) writes:
> We are having a problem with our SCO NFS package.  It seems that when
> we start doing large amounts of NFS work, the NFS and the TCP/IP just
> simply dies. There are streams resources available when this happens.

This reminds me of a question I've been meaning to ask.  Under Xenix, there
is a program called "sw" which does a really nice job of reporting, in
real-time, the Streams resources in use.  Is there an equivalent funtion
under SCO Unix?  We have all the parameters really high on one system, since
it is heavily used, but my guess is that some of them are too high, and that
there are a lot of unused resources that could be returned to user space.

Any ideas?
------------- 
James B. O'Connor			jim@tiamat.fsc.com
Ahlstrom Filtration, Inc.		615/821-4022 x. 651