Timeout problem with File Explorer and Word .. .was Re: [Samba] Time
out tuning ?
david.j as
davidj.as at gmail.com
Fri Mar 4 22:08:28 GMT 2005
This morning the same user experienced another timeout, however, this
time the Samba log showed a panic. What is very weird about this is
that the date goes into the year 1952 during the error message.
The user is still complaining that if he leaves Word open for an
extended amount of time or File Explorer, he will be disconnected and
have his Word documents will be corrupted.
This is Samba version 3.0.2-6.3E on RHEL 3.0 U1
[2005/03/04 09:52:16, 2] smbd/open.c:open_file(240)
randy opened file Program Management/eLF Accounts/eLF_Pre_Release_Accounts.xls
read=Yes write=Yes (numopen=2)
[2005/03/04 09:52:17, 2] smbd/close.c:close_normal_file(228)
randy closed file Program Management/eLF Accounts/eLF_Pre_Release_Accounts.xls
(numopen=1)
[2005/03/04 09:52:17, 2] smbd/open.c:open_file(240)
randy opened file Program Management/eLF Accounts/eLF_Pre_Release_Accounts.xls
read=Yes write=No (numopen=2)
[2005/03/04 09:52:18, 2] smbd/close.c:close_normal_file(228)
randy closed file Program Management/eLF Accounts/eLF_Pre_Release_Accounts.xls
(numopen=1)
[2005/03/04 09:53:09, 0] lib/util_sock.c:read_socket_data(342)
read_socket_data: recv failure for 4. Error = Connection timed out
[2005/03/04 09:53:09, 2] smbd/server.c:exit_server(558)
Closing connections
[2005/03/04 09:53:09, 1] smbd/service.c:close_cnum(887)
randy-vm-winxp (192.168.12.25) closed connection to service ssend_sales_share
[2005/03/04 09:53:09, 2] smbd/close.c:close_normal_file(228)
randy closed file Program Management/eLF Accounts/eLF_Pre_Release_Accounts.xls
(numopen=1)
[2005/03/04 09:53:09, 1] smbd/service.c:close_cnum(887)
randy-vm-winxp (192.168.12.25) closed connection to service randy
[2005/03/04 09:53:09, 2] smbd/close.c:close_normal_file(228)
randy closed file ssend_rlp/eLF_TdmIntro.doc (numopen=1)
[2005/03/04 09:53:09, 2] smbd/close.c:close_normal_file(228)
randy closed file ssend_rlp/~WRL0005.tmp (numopen=0)
[2005/03/04 09:59:46, 1] smbd/service.c:close_cnum(887)
randy-vm-winxp (192.168.12.25) closed connection to service docs_pending
[2005/03/04 09:59:46, 1] smbd/service.c:close_cnum(887)
randy-vm-winxp (192.168.12.25) closed connection to service n_drive
[1952/12/21 08:03:20, 0] lib/fault.c:fault_report(36)
===============================================================
[1952/12/21 08:03:20, 0] lib/fault.c:fault_report(37)
INTERNAL ERROR: Signal 11 in pid 29261 (3.0.2-6.3E)
Please read the appendix Bugs of the Samba HOWTO collection
[1952/12/21 08:03:20, 0] lib/fault.c:fault_report(39)
===============================================================
[1952/12/21 08:03:20, 0] lib/util.c:smb_panic(1422)
PANIC: internal error
[1952/12/21 08:03:20, 0] lib/util.c:smb_panic(1430)
BACKTRACE: 17 stack frames:
#0 smbd(smb_panic+0x13f) [0x8df08f]
#1 smbd [0x8cab48]
#2 /lib/tls/libc.so.6 [0x187c08]
#3 /lib/tls/libc.so.6(malloc+0x8d) [0x1d0b0d]
#4 smbd(talloc+0x3f) [0x8e475f]
#5 smbd(talloc_memdup+0x2c) [0x8e4a1c]
#6 smbd(talloc_strdup+0x3a) [0x8e4a8a]
#7 smbd [0x76f7a8]
#8 smbd(lp_logfile+0x2c) [0x76f90c]
#9 smbd(reopen_logs+0x1f6) [0x8ca226]
#10 smbd(reload_services+0x2c) [0x95581c]
#11 smbd(check_reload+0x4f) [0x7c9d2f]
#12 smbd [0x7c9dd5]
#13 smbd(smbd_process+0x13e) [0x7ca37e]
#14 smbd(main+0x505) [0x956345]
#15 /lib/tls/libc.so.6(__libc_start_main+0xf8) [0x175748]
#16 smbd [0x76ea32]
[2005/03/04 10:25:30, 2] smbd/server.c:exit_server(558)
Closing connections
[2005/03/04 10:27:40, 2] smbd/process.c:timeout_processing(1138)
Closing idle connection
[2005/03/04 10:27:40, 2] smbd/server.c:exit_server(558)
On Mon, 28 Feb 2005 10:09:50 -0800, david.j as <davidj.as at gmail.com> wrote:
> From: "Mitch (WebCob)" <mitch at webcob.com>
> To: samba at lists.samba.org
> Date: Mon, 28 Feb 2005 09:19:31 -0800
> Subject: RE: [Samba] Time out tuning ?
> > Hello,
> >
> > I have a samba server running at a customer site and they have the
> > follwing problem:
> > Employees of that company have for extended periods sometimes
> > word documents opened for editing, but it happens regularly that
> > for an half hour or more no activity takes place because they are
> > working on something else (it's a flower seed growing company , with pc's
> > in the glasshouses and they have to monitor the seedbeds and fill in
> > now and then something in in word documents). When that happens
> > the word application returns a message that , "There was a sharing
> > violation"
> ,.....
> > Is this caused by a time out on the server? or the client? Is it possible
> > to configure the samba server to forcefully keep the connection between
> > the
> > client and the server alive?
>
> >[Mitch says:] Are there any VPN's, smart switches or routers involved?
>
> >I've not seen any app timeout problems with my servers, but I know that some
> >routers or VPN arrangements have short timeouts / lack keep alive and can
> >cause this sort of behaviour.
>
> We have this same exact problem in an networked environment where
> clients are connected to a Samba server that is hosting to multiple
> subnets. I've been trying to debug this for over a year without
> success. It seems that if one user has a word document open for an
> extended amount of time Windows will corrupt the file. Also, if file
> explorer is open on a share for an extended amount of time it will
> show a disconnected dialog box. The network switches we have here in
> the network are configured to forward broadcast packets from our main
> server to all other subnets and samba is configured to advertise in
> these subnets specifically, however, this doesn't seem help.
>
> Is there anything wrong with this smb.conf file that might be causing this?
>
> [global]
>
> workgroup = NDGENG
> server string = B2 Samba
> log file = /var/log/samba/log.%m
> max log size = 50
>
> log level = 3
>
> security = user
>
> encrypt passwords = yes
> smb passwd file = /etc/samba/smbpasswd
> unix password sync = Yes
> passwd program = /usr/bin/passwd %u
> min password length = 8
> passwd chat = *New*password* %n\n *Retype*new*password* %n\n
> *passwd:*all*authentication*tokens*updated*successfully*
> username map = /etc/samba/smbusers
>
> remote announce = 192.168.3.255 192.168.7.255 192.168.11.255
> 192.168.15.255
>
> ; os level = 255
>
> socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192
> SO_KEEPALIVE IPTOS_LOWDELAY
>
> preferred master = yes
> domain master = yes
> ; local master = yes
>
> wins support = yes
>
> ; wins server = 192.168.0.4
>
> time server = no
> dns proxy = yes
>
> preserve case = yes
> short preserve case = yes
>
> default case = lower
>
> ; try to fix "lease broken - owner pid = ????" syslogs (JB 10/27/04)
> ; http://www.spinics.net/lists/samba/msg27983.html
>
> kernel oplocks = false
>
> ; try to fix file corruption (TW 09/01/04)
>
> oplock break wait time = 50
> lock spin count = 50
> lock spin time = 30
>
> debug level = 2
>
> [homes]
> comment = Home Directories
> browseable = yes
> read only = no
> create mode = 0700
>
More information about the samba
mailing list