Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!csn!boulder!daemon
From: C.Chaundy@its.unimelb.edu.au
Newsgroups: comp.dcom.sys.cisco
Subject: Trouble booting diskless Novell clients across a router
Message-ID: <34768@boulder.Colorado.EDU>
Date: 6 May 91 15:06:18 GMT
Sender: daemon@boulder.Colorado.EDU
Lines: 30

We are having difficulty in booting diskless Novell clients across a
cisco AGS+ router (running 8.2-3 code).  The clients are all on one
Ethernet interface and the Novell server is on another.

This is the problem analysis as it was reported to me:

Novell keeps track of all the Ethernet addresses of its workstations by
giving them a connection number when they first connect to the server
and then it remembers this number until you shutdown the server.

The problem with the cisco showed up when we were using remote boot eproms
in our ethernet cards (Micom NI5210's).  Novell would begin to boot the
machine (by downloading a psuedo-boot-disk), but when it came to loading
the true network driver, the machine would hang.  What was happening was
that the Server was giving the machine a new Connection number at this time
and then the psuedo-boot-disk would lose track of where the machine was.
This was also causing the server to lock the psuedo-boot-disk file while
the "lost" connection was still trying to "read" it thus not allowing it
to be modified in any way.

Is this analysis correct and how can the situation be rectified?

Thanks in advance, Chris Chaundy

Technical Manager, Networks,
Information Technology Services, The University of Melbourne
Internet: C.Chaundy@its.unimelb.EDU.AU	(DTE 505233430003)
Phone:    +61 3 344 7045  Cables Unimelb
Fax:      +61 3 347 4803  Telex AA35185
Post:     Parkville, Victoria 3052 Australia