Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 23 Nov 2002 16:38:13 -0800
From:      "David Smithson" <david@customfilmeffects.com>
To:        <stacey@vickiandstacey.com>
Cc:        <questions@FreeBSD.ORG>
Subject:   Re: Network connection dropping during file transfers
Message-ID:  <00bf01c29351$c649baf0$0801a8c0@customfilmeffects.com>
References:  <007101c2934a$8bad2b40$0801a8c0@customfilmeffects.com> <00a101c2934d$bfb43b60$0801a8c0@customfilmeffects.com> <1038097743.43256.45.camel@localhost>

next in thread | previous in thread | raw e-mail | index | archive | help
> Hi David,
>    By chance, is it possible for you to let the list know:-

Yes.

> 1]Client OS

Clients are Windows 2000 Professional SP3 and Windows 2000 Server SP3.

> 2] Are you using DHCP?

Yes.  A Windows 2000 domain controller also provides DHCP services to the
Windows 2000 Pro clients.  All servers are static.

> 3] Are you using a Wins Server for your domain?

The domain controller currently provides WINS and DNS services.

> 4] For how long does the connection stay off-line before coming back?

This period seems to be indefinite based on unquantified observations.  I
have not timed this period.  A few times it seemed never to come back up.  I
even tried ifconfig nge0 down && ifconfig nge0 up to re-initialize the
interface.  This didn't help.  Rebooting the server fixed the interface in
these instances.

> The reasoning behind the above is the fact that this might an issue of
> how (if this is so) MS Windows clients on dhcp maintain respective
> connections over the network during data transfer - You may well be
> seeing cases of timeouts on the client end here.

I'll do whatever it takes to get this working.  Anything else I should look
into?

> Stacey
>
>
> On Sun, 2002-11-24 at 00:09, David Smithson wrote:
> > P.S.  I'm using Samba 2.2.7.
> >
> >
> > ----- Original Message -----
> > From: "David Smithson" <david@customfilmeffects.com>
> > To: <questions@FreeBSD.ORG>
> > Sent: Saturday, November 23, 2002 3:46 PM
> > Subject: Network connection dropping during file transfers
> >
> >
> > > Hi.  I'm having trouble with a mission-critical file-server.  Server
is
> > > FreeBSD 4.5, exporting the filesystem with Samba over gigabit copper.
> > > Clients are Windows 2000.  During multiple read/write operations
between
> > the
> > > clients and the server, the network connection to the server will drop
out
> > > for an indefinite period of time and then somehow reconnect.  This
> > obviously
> > > interrupts file transfers and foils all of the Windows clients'
> > > explorers.exe processes.  This also leads to incomplete files on the
> > server
> > > and frozen smbd processes.  Dmesg outputs nothing of interest.
Smbstatus
> > > outputs this:
> > >
> > > Samba version 2.2.2
> > > Service      uid      gid      pid     machine
> > > ----------------------------------------------
> > > W            nobody   wheel     1357   bunnicula (192.168.1.5) Sat Nov
23
> > > 15:27:19 2002
> > > X            nobody   wheel     1357   bunnicula (192.168.1.5) Sat Nov
23
> > > 15:27:19 2002
> > > W            nobody   wheel     1308   render-02 (192.168.1.12) Sat
Nov 23
> > > 15:17:00 2002
> > > W            nobody   wheel     1312   render-01 (192.168.1.10) Sat
Nov 23
> > > 15:17:09 2002
> > > W            nobody   wheel     1318   ethel    (192.168.1.8) Sat Nov
23
> > > 15:17:36 2002
> > > W            shaina   wheel     1326   lucy     (192.168.1.15) Sat Nov
23
> > > 15:18:09 2002
> > > W            amani    wheel     1283   bunnicula (192.168.1.5) Sat Nov
23
> > > 15:10:26 2002
> > > X            amani    wheel     1283   bunnicula (192.168.1.5) Sat Nov
23
> > > 15:10:30 2002
> > > X            shaina   wheel     1326   lucy     (192.168.1.15) Sat Nov
23
> > > 15:21:09 2002
> > > W            nobody   wheel     1311   ricky-athlon (192.168.1.13) Sat
Nov
> > > 23 15:17:08 2002
> > > W            nobody   wheel     1309   render-03 (192.168.1.11) Sat
Nov 23
> > > 15:17:04 2002
> > > W            administrator wheel     1284   render-01 (192.168.1.10)
Sat
> > Nov
> > > 23 15:10:26 2002
> > > X            administrator wheel     1284   render-01 (192.168.1.10)
Sat
> > Nov
> > > 23 15:10:33 2002
> > >
> > > Locked files:
> > > Pid    DenyMode   R/W        Oplock           Name
> > > --------------------------------------------------
> > > 1284   DENY_ALL   WRONLY     EXCLUSIVE+BATCH
> > > /array1/renders/Daredevil/film/OP34-02/OP34-02_0118.cin   Sat Nov 23
> > > 15:10:53 2002
> > > 1283   DENY_ALL   WRONLY     EXCLUSIVE+BATCH
> > > /array1/renders/Daredevil/film/OP34-02/OP34-02_0119.cin   Sat Nov 23
> > > 15:10:54 2002
> > >
> > > The two processes above (1284 and 1283) don't release their locks
until
> > they
> > > are killed.
> > >
> > > log.smbd | tail -n 50 shows this:
> > >
> > >   write_socket_data: write failure. Error = Broken pipe
> > > [2002/11/23 12:19:13, 0] lib/util_sock.c:write_socket(566)
> > >   write_socket: Error writing 65534 bytes to socket 12: ERRNO = Broken
> > pipe
> > > [2002/11/23 12:19:13, 0] lib/util_sock.c:send_smb(730)
> > >   Error writing 65534 bytes to client. -1. (Broken pipe)
> > > [2002/11/23 12:24:20, 0] smbd/nttrans.c:call_nt_transact_ioctl(1762)
> > >   call_nt_transact_ioctl: Currently not implemented.
> > > [2002/11/23 12:32:07, 0] lib/util_sock.c:open_socket_in(830)
> > >   bind failed on port 139 socket_addr = 192.168.1.24.
> > >   Error = Can't assign requested address
> > > [2002/11/23 13:05:23, 0] lib/util_sock.c:write_socket_data(542)
> > >   write_socket_data: write failure. Error = Broken pipe
> > > [2002/11/23 13:05:23, 0] lib/util_sock.c:write_socket(566)
> > >   write_socket: Error writing 65534 bytes to socket 12: ERRNO = Broken
> > pipe
> > > [2002/11/23 13:05:23, 0] lib/util_sock.c:send_smb(730)
> > >   Error writing 65534 bytes to client. -1. (Broken pipe)
> > > [2002/11/23 13:06:17, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 13:07:00, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 13:07:04, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 13:07:06, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 15:11:25, 0] lib/util_sock.c:read_socket_with_timeout(300)
> > >   read_socket_with_timeout: timeout read. read error = Connection
reset by
> > > peer.
> > > [2002/11/23 15:11:26, 0] smbd/oplock.c:oplock_break(782)
> > >   oplock_break: receive_smb error (Connection reset by peer)
> > >   oplock_break failed for file Scans
> > > In/daredevil/dd_op34_02_c34c_3/dd_op34_02_c34c_3.0110.cin (dev =
29300,
> > > inode = 1063922).
> > > [2002/11/23 15:11:26, 0] smbd/oplock.c:oplock_break(870)
> > >   oplock_break: client failure in break - shutting down this smbd.
> > > [2002/11/23 15:11:55, 0] smbd/oplock.c:request_oplock_break(1026)
> > >   request_oplock_break: no response received to oplock break request
to
> > pid
> > > 1285 on port 1371 for dev = 29300, inode = 1063922
> > >   for dev = 29300, inode = 1063922, tv_sec = 3de00afe, tv_usec = 15342
> > > [2002/11/23 15:17:00, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 15:17:00, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 15:17:09, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 15:17:23, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 15:17:45, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 15:19:05, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 15:19:11, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > > [2002/11/23 15:19:37, 0] lib/util_sock.c:read_socket_data(478)
> > >   read_socket_data: recv failure for 4. Error = Connection reset by
peer
> > >
> > >
> > > What does this indicate?  Have I provided enough information here?
> > >
> > > By the way, the transfers that are taking place are initiated by a
visual
> > > effects / compositing program called Digital Fusion.  The clients are
> > render
> > > engines that are currently being used to render shots for the upcoming
> > > feature film "Chicago".  I must deliver these rendered frames tomorrow
by
> > 9
> > > p.m.  :(  Your help will be much appreciated.
> > >
> > > --
> > > David Smithson - Systems Administrator
> > > Custom Film Effects (http://www.customfilmeffects.com)
> > >
> > >
> > >
> > >
> > >
> > > To Unsubscribe: send mail to majordomo@FreeBSD.org
> > > with "unsubscribe freebsd-questions" in the body of the message
> >
> >
> >
> > To Unsubscribe: send mail to majordomo@FreeBSD.org
> > with "unsubscribe freebsd-questions" in the body of the message
> --
> Stacey Roberts
> B.Sc (HONS) Computer Science
>
> Web: www.vickiandstacey.com
>



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?00bf01c29351$c649baf0$0801a8c0>