smbd crash and/or not running

[olympia]

A wasn't able to find out what is the triggering the crash of smb (most of the times it happens when I try delete something during file transfers), but the point is, that smbd stops running.
When I log in with ssh, and do a "ps -ef | grep mbd", I only see nbmd running, but not smbd.
Rebooting is not solving the problem, only if I shutdown the whole thing and power up again".

Does anybody have any clue how can I troubleshoot this issue?
I was trying to find some log regarding details about the failure, but all the smb logs are empty...

Is it possible that is relating to a hardware failure?

[dbridges]

olympia said:

A wasn't able to find out what is the triggering the crash of smb (most of the times it happens when I try delete something during file transfers), but the point is, that smbd stops running.
When I log in with ssh, and do a "ps -ef | grep mbd", I only see nbmd running, but not smbd.
Rebooting is not solving the problem, only if I shutdown the whole thing and power up again".

Does anybody have any clue how can I troubleshoot this issue?
I was trying to find some log regarding details about the failure, but all the smb logs are empty...

Is it possible that is relating to a hardware failure?

You could try starting it manually.

have a look at /app/etc/rc.d/rc.samba

N2100:/app/etc/rc.d# cat rc.samba
#!/bin/sh
#
# /etc/rc.d/rc.samba
#
# Start/stop/restart the Samba SMB file/print server.
#
# To make Samba start automatically at boot, make this
# file executable: chmod 755 /etc/rc.d/rc.samba
#
smbhome=/opt/samba
#smbhome=/usr
samba_start() {
if [ -x ${smbhome}/sbin/smbd -a -x ${smbhome}/sbin/nmbd -a -r /etc/samba/smb.conf ]; then
echo "Starting Samba: ${smbhome}/sbin/smbd"
${smbhome}/sbin/smbd
echo " ${smbhome}/sbin/nmbd"
${smbhome}/sbin/nmbd
fi
}

samba_stop() {
killall smbd nmbd
}

samba_restart() {
samba_stop
sleep 2
samba_start
}

case "$1" in
'start')
samba_start
;;
'stop')
samba_stop
;;
'restart')
samba_restart
;;
*)
# Default is "start", for backwards compatibility with previous
# Slackware versions. This may change to a 'usage' error someday.
samba_start
esac

N2100:/app/etc/rc.d#

[olympia]

Yes, I tried, unfortunetly with the same result...
I increased the log level in samba config, than I get the below error in samba log:

[2007/03/21 02:14:24, 2] lib/interface.c:add_interface(81)
added interface ip=192.168.0.2 bcast=192.168.0.255 nmask=255.255.255.0
[2007/03/21 02:14:24, 2] lib/interface.c:add_interface(81)
added interface ip=192.168.2.100 bcast=192.168.2.255 nmask=255.255.255.0
[2007/03/21 02:14:24, 3] lib/util.c:interpret_addr(1273)
sys_gethostbyname: Unknown host. wlan*
[2007/03/21 02:14:24, 2] lib/interface.c:interpret_interface(130)
can't determine netmask for wlan*
[2007/03/21 02:14:24, 2] lib/interface.c:add_interface(81)
added interface ip=127.0.0.1 bcast=127.255.255.255 nmask=255.0.0.0
[2007/03/21 02:14:24, 3] smbd/server.c:main(877)
loaded services
[2007/03/21 02:14:24, 0] smbd/server.c:main(881)
standard input is not a socket, assuming -D option
[2007/03/21 02:14:24, 3] smbd/server.c:main(892)
Becoming a daemon.
[2007/03/21 02:14:24, 2] lib/tallocmsg.c:register_msg_pool_usage(61)
Registered MSG_REQ_POOL_USAGE
[2007/03/21 02:14:24, 2] lib/dmallocmsg.c:register_dmalloc_msgs(71)
Registered MSG_REQ_DMALLOC_MARK and LOG_CHANGED
[2007/03/21 02:14:24, 0] lib/fault.c:fault_report(41)
===============================================================
[2007/03/21 02:14:24, 0] lib/fault.c:fault_report(42)
INTERNAL ERROR: Signal 11 in pid 2766 (3.0.23a)
Please read the Trouble-Shooting section of the Samba3-HOWTO
[2007/03/21 02:14:24, 0] lib/fault.c:fault_report(44)

From: www.samba.org/samba/docs/Samba3-HOWTO.pdf
[2007/03/21 02:14:24, 0] lib/fault.c:fault_report(45)
===============================================================
[2007/03/21 02:14:24, 0] lib/util.c:smb_panic(1592)
PANIC (pid 11879879540736): internal error
[2007/03/21 02:14:24, 0] lib/util.c:log_stack_trace(1699)
BACKTRACE: 1 stack frames:
#0 /opt/samba/sbin/smbd [0x435d7c]
[2007/03/21 02:14:24, 0] lib/fault.c:dump_core(168)
unable to change to /opt/samba/var/cores/smbdrefusing to dump core


I assume there has to be some hw failure...

[dbridges]

olympia said:

I assume there has to be some hw failure...

Probably more of a hardware limitation...

How much memory have you got?


I'll be honest and sat that I have very little idea about how to resolve this. Sorry.

[olympia]

It is 256Megz. Do you think that is not enough to handle the traffic?

However, for testing purposes I just tried to transfer 3 large files to the box and read one from the box at the same time. Result: tho whole device crashed after a while. I was not able to access it any way, neither I was able to use the power button to power off, so I had to plug out the power cable. Than after rebooting I realised that the raid was crashed as well and it is recovering itself.

Would it really be a hw limitation and it is "normal"?
But this device is build as a NAS, this is what is should able to do, nothing more...

[olympia]

After a few days of testing, finally I can say, that the root cause of my long nights was the faulty RAM modul. I replaced it, and it works smoothly for days up until now.

dbridges, thanks for focusing my attention to the RAM, you was the one, who put the bug in my ears

ps. I found out that cause a lot of corrupted files on the drives as well

[dbridges]

Glad it's working.

I must admit that when i upgraded from the 128MB to 512MB that i was really worried. I made sure that i bought the same make and model as the original with the same timing figures.

Your ram may not be faulty. It may just be incompatible.

[olympia]

It was the original modul, as my device is an Allnet version, and Allnet is shiping it with 256MB by default.

So, it is faulty, by factory, but I don't blame, now it's OK.