SUMMARY: 3.2C NFS servers

From: James D. Zelenka <jz1j+_at_andrew.cmu.edu>
Date: Wed, 7 Feb 1996 11:12:20 -0500 (EST)

The question:
>We have a bunch of alphas here that were acting as NFS servers while
>running 3.2. With a recent upgrade to 3.2C, clients have become unable to
>operate reasonably upon filesystems mounted from these servers. The
>mount still succeeds, and the client can "df" the filesystem, but any
>attempts to ls, read, or write a file are doomed to hang. This includes
>3.2C and 3.2 clients running on other alphas, along with some rs/6000s,
>suns, HPs, and other workstations.

>Has anyone seen or heard of this before? Any known workarounds? Any help
>would be _greatly_ appreciated!

Summary of responses:

"Mike DeMarco <MDEMARCO.US.ORACLE.COM>" <MDEMARCO_at_us.oracle.com>

> I believe that there is a patch for this type of behaviour that
> you are describing. See the attached.

Attached was the description of patches:

PROBLEM 1: (Patch ID: OSF320-104) (HPXQ54CA7,HPXQ57F64)
(fixes a memory leak caused by AIX/SUN and possibly other clients accessing
DU NFS server)
PROBLEM 2: (Patch ID: OSF320-126) (HPAQ6190F)
(fixes problems handling reused transaction ids)

---
It turns out that the fix to problem I had was handled by:
PROBLEM:	(Patch ID: OSF350-109) 		(Case ID: MGO101612)
Which is the 3.2C version of problem 2 above (it's a patched svc_kudp.o)
-Jim Zelenka
Received on Wed Feb 07 1996 - 17:43:57 NZDT

This archive was generated by hypermail 2.4.0 : Wed Nov 08 2023 - 11:53:46 NZDT