lkml.org 
[lkml]   [1999]   [Mar]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: SMBfs problems
Markus Fischer wrote:
>
> i sadly have to commit that i have the same problem running two linux
> machines (server: linux 2.2.2/samba 2.0.3; client linux 2.2.3).
> but it doesnt apply to heavy load. more it happens after a long time
> ( >16 hours or so). i tried with 2.0.36 for the client and this
> one worked fine. so i also guess its some smbfs trouble in here ?
>
> markus
>
> Once upon a time (around Tue, Mar 23, 1999 at 11:56:22PM -0800 i think)
> Russ Steffen told me :
> >
> > I'm having a rather curious problem with some NT shares that I have
> > mounted via SMBfs.
> >
> > The shares mount fine, and work for well under "light" loads, but when
> > under stress (like a BRU or tar backup, find, or anything that does a lot
> > of directory scanning) the share "disappears" (ie. the top directory
> > suddenly appears empty save for . and .. ). The share is still mounted
> > according to mount, and df still reports the correct capacity and usage.
> > The only way I've found to recover is to umount & remount it.
> >
> > The syslog is absolutely silent on the subject the only oddities I've
> > found are a couple of messages like:
> >
> > Mar 15 17:31:49 saturn kernel: smb_dont_catch_keepalive:
> > server->data_ready == NULL
> >
> > - and -
> >
> > Mar 19 19:51:52 saturn kernel: smb_statfs: dskattr error = 5
> >
> > However, they don't correspond to the event. They appear at a times when
> > the mount was healthy and there is no error logged at the time of death.
> >
> > Other relevant info:
> > Kernel 2.0.36, observed on both SMP and UP.
> >
> > I've observed this on shares mounted from both NT Server 3.51 SP5
> > and 4.0 SP3.
> >
> > smbclient (via smbtar) has no problems on these shares, it's just smbfs.
> >
> > Russ
> > ----------------------------------
> > Russ Steffen
> > rsteffen@ia.net

I've also had similar problems, using kernel 2.2.x/samba 2.0.2,
connecting to NT 4.0sp4 WS. Usually, I get an IO error, or something
similar, which according to the docs was attributed to a keepalive
error. Also, for me, the problem is only one-way. Going from Linux to
NT, the smb share times out, usually fairly quickly (less than 1/2
hour). Going from NT to Linux, it works fine. Don't know if that last
matters, though.

Hmm, /var/log/messages has a strikingly similar error message to the one
above:
Mar 1 08:43:03 reliant kernel: smb_dont_catch_keepalive:
server->data_ready == NULL


I can see where this is defined in smbfs/sock.c, but what conditions can
cause this error to be generated? When we can determine that, we should
be able to fix the problem.

Of course, that assumes that the problem is with the kernel and not with
samba 2.0.x.
--
Matthew Vanecek
Studies in Business Computers at the University of North Texas
http://www.unt.edu/bcis
*****************************************************************
Visit my Website at http://people.unt.edu/~mev0003
*****************************************************************
For 93 million miles, there is nothing between the sun and my shadow
except me. I'm always getting in the way of something...
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:51    [from the cache]
©2003-2011 Jasper Spaans