lkml.org 
[lkml]   [1997]   [Jul]   [30]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: NFS troubles when mounting multihomed server
In article <Pine.LNX.3.96.970723063614.4255F-100000@moisil.cs.columbia.edu> you wrote:
: On Wed, 23 Jul 1997, Swen Thuemmler wrote:

: > You should check the version of /sbin/mount - newer versions don't connect
: > the socket (if your kernel is not too old (below 1.1.3x, I think)). And
: > you might check the version of amd, if you use automounting. The debian
: > version contains the neccesa^H^H^H^H^Hsseca^H^H^H^H^H^H^Hneeded (:-)
: > patches.
: >
: > Hope this helps.

: It does, thanks a lot. And shame on me for not mentioning the kernel
: version - it's pre2.0.31-2 but the problem applies to any recent kernel.

: I do have the latest mount, but based on my discution with amd's
: maintainer I didn't think the problem was in mount.. now I know better :)

: [nfsmount.c]
: * Fri, 8 Mar 1996 18:01:39, Swen Thuemmler <swen@uni-paderborn.de>:
: * Omit the call to connect() for Linux version 1.3.11 or later.

: I was testing the alpha version of amd 6.0 which does not have the debian
: patches yet. The maintainer asked me if there is any flag similar to
: NetBSD's NFSMNT_NOCONN to avoid connecting the socket, I didn't find
: anything so I went through the nfs code to find some clues. Then posted
: the message..

: Ok, so I referred amd's maintainer to the debian patches, the relevant
: portions will be incorporated in the next alpha version of am-utils 6.0.

: Thanks again,
: Ionut

I have been talking with H.J. Lu about problems I have been having with amd
and NIS under libc 5.4.33. I am using the amd 102 -11 from debian too.

Basically, under .33, amd only works
a) using a file for its map
or b) using a NIS page served from a YP server on the same machine.

i.e. - maps don't distribute via NIS. H.J. says that NIS is broken or buggy in
5.4.33 and has asked me to try 5.4.36 instead. This is all with the right (HJ)
ypclients, and 1.2 ypserv. 5.4.36 is better in some respects, but I have had
to retreat to 5.4.33 for zlib compatibility and other issues. machines running
under libc 5.4.36 and 5.4.33 do not mix via NIS.

I would like amd to run off a map served from my solaris server (NIS+ running
in NIS compatibility mode), but yp_order seems unsupported under NIS+ and amd
relies heavily on it. I would wish for the amd code to be rewritten to avoid
yp_order() calls ... I can't push maps from solaris to linux NIS servers for
the same kind of reasons.

Basically, if anybody has amd working via NIS, I would like to know how.
I can just about manage it using an intricate and delicate sequence of
yppull's and NIS re-exports to ypserv's feeding only loopback nets, but, that
is not right.

: --
: It is better to keep your mouth shut and be thought a fool,
: than to open it and remove all doubt.



--
---------------------------------------------------------------------
Peter T. Breuer Phd. Ing.,
Area de Ingenieria Telematica E-mail: ptb@it.uc3m.es
Dpto. Ingenieria Tel: +34 1 624 99 53
Universidad Carlos III de Madrid Fax: +34 1 624 94 30/65
Butarque 15, E-28911 Leganes URL: http://www.it.uc3m.es/~ptb
Spain

\
 
 \ /
  Last update: 2005-03-22 13:40    [W:0.037 / U:0.680 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site