Re: Tr:Re: cifs oplock windows share

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, 05 Mar 2012 14:53:35 +0100
"sergio.conrad" <sergio.conrad@xxxxxxxxxxx> wrote:

> > 
> > 4°) Everything is working fine when working only with a few workstations.
> > 
> > 5°) But when the students were working on severals, we find some problems.
> > In the /var/log/messages
> > Oct 5 16:42:24 u1209-01l kernel: [ 58.783078] CIFS VFS: No response for cmd 114 mid 1
> > Oct 5 16:42:24 u1209-01l kernel: [ 58.783094] CIFS VFS: cifs_mount failed w/return code = -112
> > Oct 5 16:42:24 u1209-01l kdm: :0[1926]: pam_mount(mount.c:64): Errors from underlying mount program:
> > Oct 5 16:42:24 u1209-01l kdm: :0[1926]: pam_mount(mount.c:68): mount error(112): Host is down
> > Oct 5 16:42:24 u1209-01l kdm: :0[1926]: pam_mount(mount.c:68): Refer to the mount.cifs(8) manual page (e.g. man mount.cifs)
> > Oct 5 16:42:24 u1209-01l kdm: :0[1926]: pam_mount(pam_mount.c:521): mount of data/u20 failed
> > The windows 2008 server share hangs up and must be rebooted
> > 

Most of the other data in this isn't terribly useful, but the above
tells us a little. cmd 114 is a NEGOTIATE call and is the first call
sent on any socket connection to the server. If the server isn't
responding to that then it sounds like something is very wrong there.

It's possible that the problem there is triggered by load generated by
the clients. Recent kernel updates have converted cifs to use
asynchronous writes for better performance so it's possible that is
exacerbating your problem somehow. We don't really have a way to know
though. Either way, this really looks like a problem on the server side.

One thing you might experiment with is reducing the setting of the
cifs_max_pending module parameter on your clients. That will reduce the
number of calls that they are allowed to have in flight to the server
at any particular time. That's just speculation though...

The right solution though is to contact the server vendor (MS in this
case) and see if they might be able to help diagnose why the server
isn't responding in a timely fashion.

-- 
Jeff Layton <jlayton@xxxxxxxxx>
--
To unsubscribe from this list: send the line "unsubscribe linux-cifs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux