HP3000-L Archives

June 1999, Week 3

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:
Sletten Kenneth W KPWA <[log in to unmask]>
Reply To:
Sletten Kenneth W KPWA <[log in to unmask]>
Date:
Thu, 17 Jun 1999 18:11:09 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (57 lines)
Donna flogs that slow FTP dog, after Mike complained:

> > FTP is running abysmally slow.

> mike -- i've run into the same/similar problem but no, i don't
> have any answers....

> i'm trying to send files from my 3000 to an nt *across* the internet
> and having a dreadful time getting it to work.  i opened a call to hp ....
> imo, hp became less helpful :-(  they're suggestion was to bringing
> hp in on a consulting basis to examine and *possibly* adjust our
> network configurations.  ....

From my admittedly limited experience, the amount of system-
level network "help" you can get from the HP RC seems to
depend to a large degree on the luck of the draw of which RC
engineer you happen to get:  A long time ago (year +) I had a
chronic network problem of some kind, and while the RC person
did look into it to some degree they also said that an "in depth"
analysis would bring up the dreaded "consulting fee" situation.

Then just over the last couple weeks I got an extremely helpful
RC engineer, who even from my jaded perspective went above
and beyond the call:  We have been suffering for an extended
period from some kind of asynchronous and complex network
"events";  where connections to our 959KS/200 going through
the router(s) will just freeze from anywhere between 10-15 sec
to 30-40 minutes (our 3000 is not the only affected host);  and
then just magically resume working again.  The local network
techies say "we can't see anything wrong".  Meanwhile, direct
network connections via the switching hubs continue with no
discernable problem.....   arrggghhhh.....

Anyway, from various monitoring and investigation that I've been
doing I'm pretty well convinced the 3000 is not the cause or even
a significant contributor to the problem.  I even mentioned that to
the RC engineer.  She still took the time to go through every last
setting in NMMGR for all three of our configured networks, and
made a couple of recommendations for changes to optimize
operation in our environment.....  didn't solve our "connection thru
the routers" problem (didn't expect it to), but it was nice that she
took the time to do the drill....


SIDEBAR:   The above helpful RC engineer said the manual is
a bit out of date WRT recommended settings in NMMGR in some
cases.  Five years ago the average site's network traffic was
"simpler and cleaner" than today.  One recommendation:  On the
NETXPORT.NI.LAN46.PROTOCOL.ARP  screen:

For networks that are fairly heavily loaded;  and especially those
that have a "variety" of traffic with at least occasional maximum
bursts:  Set the "Retransmission Timeout"  up to 3 seconds,
instead of the earlier recommended 1 second....

Ken Sletten

ATOM RSS1 RSS2