HP3000-L Archives

August 2000, Week 4

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Donna Garverick <[log in to unmask]>
Reply To:
Donna Garverick <[log in to unmask]>
Date:
Fri, 25 Aug 2000 10:05:32 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (28 lines)
Bruce Toback wrote:

> Have you tried running nettool/resource?

oh yes...

> I have seen at least one case
> similar to this where the outbound buffer pool was too small. The system
> would sometimes not recover gracefully from an overflow condition.
> Perhaps you could monitor network resource usage closely and see if
> you're running up against one limit or another.

so far we've not seen anything unusual via nettool.  we did see something
interesting yesterday once we put a sniffer on.  we were pinging from one of
the two 'problem' mpe systems to the router.  the ping (from the 3000) would
eventually stall yet on the sniffer you could see that packets were still
being received and answered.  in the words of our sr. network engineer: 'it
appears that LAYER 2 (MAC) communications continues until the target number of
frames has been transmitted.  At the application layer, communications is lost
and the total send/receive count is under reported.'             - d 'off to
call hp'

--
Donna Garverick     Sr. System Programmer
925-210-6631        [log in to unmask]

>>>MY opinions, not Longs Drug Stores'<<<

ATOM RSS1 RSS2