From: "Ulrich Windl" <ulrich.windl@rz.uni-regensburg.de>
Subject: Re: Problems with SLES10 (2.6.16.21-0.25-xen) NFS client: no
	locks, hanging mounts
Date: Fri, 17 Nov 2006 08:22:30 +0100
Message-ID: <455D7145.16120.16D333@Ulrich.Windl.rkdvmks1.ngate.uni-regensburg.de>
References: <455C515D.12136.104B8B29@Ulrich.Windl.rkdvmks1.ngate.uni-regensburg.de>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Cc: nfs@lists.sourceforge.net
Return-path: <nfs-bounces@lists.sourceforge.net>
Received: from sc8-sf-mx2-b.sourceforge.net ([10.3.1.92]
	helo=mail.sourceforge.net)
	by sc8-sf-list2-new.sourceforge.net with esmtp (Exim 4.43)
	id 1Gky4w-0001VV-CZ
	for nfs@lists.sourceforge.net; Thu, 16 Nov 2006 23:24:02 -0800
Received: from rrzmta2.rz.uni-regensburg.de ([132.199.1.17])
	by mail.sourceforge.net with esmtp (Exim 4.44) id 1Gky4u-0004D6-SD
	for nfs@lists.sourceforge.net; Thu, 16 Nov 2006 23:24:03 -0800
To: Trond Myklebust <trond.myklebust@fys.uio.no>
In-reply-to: <1163697009.9593.25.camel@lade.trondhjem.org>
List-Id: "Discussion of NFS under Linux development, interoperability,
	and testing." <nfs.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/nfs>,
	<mailto:nfs-request@lists.sourceforge.net?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=nfs>
List-Post: <mailto:nfs@lists.sourceforge.net>
List-Help: <mailto:nfs-request@lists.sourceforge.net?subject=help>
List-Subscribe: <https://lists.sourceforge.net/lists/listinfo/nfs>,
	<mailto:nfs-request@lists.sourceforge.net?subject=subscribe>
Sender: nfs-bounces@lists.sourceforge.net
Errors-To: nfs-bounces@lists.sourceforge.net

On 16 Nov 2006 at 12:10, Trond Myklebust wrote:

> On Thu, 2006-11-16 at 11:54 +0100, Ulrich Windl wrote:
> > Hi,
> > 
> > (I'm not subscribed to this list)
> > When I had a temporary need for more diskspace on a XEN VM running SLES10 
> > (x86_64), I tried to mount an NFS filesystem to Linux, but I had problems:
> > 
> > 1) When Solaris 9 (SPARC) was the server (server1), mount succeeded, but locking 
> > failed:
> > 
> > Application (SuSE "build") said:
> > initializing rpm db...
> > warning: waiting for transaction lock on /var/lib/rpm/__db.000
> > error: can't create transaction lock on /var/lib/rpm/__db.000
> > 
> > /var/log/messages said:
> > Nov 16 10:27:31 nfs-client kernel: lockd: failed to monitor server1
> > Nov 16 10:27:31 nfs-client kernel: lockd/statd: failed to create 
> > /var/lib/nfs/sm/server1: err=-2
> 
> err == -2 is the same as ENOENT.
> 
> Does the directory /var/lib/nfs/sm exist and have the proper
> permissions?

Yes (I think):
Client: drwx------ 2 root root 48 Jul  3 21:11 /var/lib/nfs/sm
Server: drwx------  2 root root 48 Nov 17  2005 /var/lib/nfs/sm

The odd thing is that when trying to _create_ a file in an existing directory, 
ENOENT is probably an unexpected error message. Isn't it?

> 
> > 2) When trying to use SLES9 (i386, Kernel 2.6.5-7.282-default) as NFS server 
> > (server2), mount on the client hung hard (only kill -9 did work). The rpc.mountd 
> > was waiting for data while the client hung:
> > 
> > strace:
> > open("/proc/fs/nfsd/filehandle", O_RDWR) = 10
> > fstat64(10, {st_mode=S_IFREG|0600, st_size=0, ...}) = 0
> > mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 
> > 0x40019000
> > write(10, "*.dvm.our-domain,r"..., 85) = 85
> > read(10, "\\x0100000600fd000b020000006e2200"..., 4096) = 75
> > close(10)                               = 0
> > munmap(0x40019000, 4096)                = 0
> > write(8, "\200\0\0Pb\376%\321\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 84) = 84
> > select(1024, [3 4 5 6 7 8], NULL, NULL, NULL) = 1 (in [8])
> > poll([{fd=8, events=POLLIN, revents=POLLIN}], 1, 35000) = 1
> > read(8, "", 4000)                       = 0
> > close(8)                                = 0
> > select(1024, [3 4 5 6 7], NULL, NULL, NULL <unfinished ...>
> > 
> > Tracing the network traffic, I found this final data exchange:
> > 
> > 11:29:43.869678 IP (tos 0x0, ttl  64, id 50629, offset 0, flags [DF], proto: TCP 
> > (6), length: 200) nfs-client.dvm.our-domain.2253585828 > server2.dvm.our-
> > domain.nfs: 148 fsinfo fh 
> > Unknown/0100000600FD000B020000006E22000002000000000000000200000001000000
> > 11:29:43.909121 IP (tos 0x0, ttl  64, id 29964, offset 0, flags [DF], proto: TCP 
> > (6), length: 52) server2.dvm.our-domain.nfs > nfs-client.dvm.our-domain.1023: ., 
> > cksum 0x655b (correct), ack 542460337 win 6432 <nop,nop,timestamp 470378359 
> > 19011966>
> > 
> > Any helpful instructions?
> 
> No idea about this one. It looks as if your server is failing to reply
> to the client's fsinfo inquiry. All it appears to have done is to ack
> the TCP message.
> 
> Is mountd up and running?

The mountd was the process being strace'd, so I guess, yes, it is running.

Regards,
Ulrich


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs