HI All.
We've been having some problems the past couple days with one of our
Samba Servers. We currently have it running two instances. One
instance is using local auth (historical) the other instances is using
our newish domain for authentication. We're slowly migrating users
over to the domain so having both methods available was the solution we
came up with.
Yesterday we had a crash and couldn't remotely login to the server. Our
data partition is VERY near capacity, but the system and var partitions
are fine. After a reboot some strange issues with the smb, nmb and
winbind processes.. couldn't stop/start them properly. kept complaining
they were already running. The domain users could access their drives,
but not print... and the non domain users couldn't access their drives,
but could print. hah.
After another reboot and a fsck, we started up the local auth instance
of samba and seemed ok. and started up the domain auth instance and
things seems to be working.
Over the course of the night about 1900 smb processes had built up and
the load avg on the server was through the roof. One user alone had
1100 of those processes.
In the logs there was a samba panic and internal error (more details
and backtrace to follow)
We're unfortunately running an older version of samba, on an older
distro of SLES... but everything was working fine for almost the past year.
SLES 10.1
kernel 2.6.16.46-0.12-smp
samba-3.0.36-7.1
smb.conf
[global]
# client code page = 437
interfaces = [ip removed].10/24
bind interfaces only = true
workgroup = WORKGROUP
netbios name = smbserver
netbios aliases = CBNsmbserver
server string = smbserver
encrypt passwords = Yes
map to guest = Bad User
passwd program = /usr/bin/passwd
log file = /var/log/samba/log.smbd
lock directory = /var/lib/samba
time server = Yes
kernel oplocks = No
oplock break wait time = 20
deadtime = 10
keepalive = 30
admin users = george
os level = 34
local master = Yes
preferred master = Yes
# domain master for WORKGROUP
domain master = Yes
# The only WINS server
wins support = Yes
# Local master browsers on the different subnets plus locale1 domain master
remote browse sync = [ips removed]
remote announce = [ip removed]/WORKGROUP [ip removed]/PHONE
unix extensions = Yes
log level = 2
max log size = 1048576
syslog = 0
printing = CUPS
printcap name = CUPS
use client driver = Yes
socket options = SO_KEEPALIVE IPTOS_LOWDELAY TCP_NODELAY
# veto files = /*.eml/*.nws/riched20.dll/*.{*}-
browseable = Yes
writeable = Yes
create mask = 0600
directory mask = 0700
locking = Yes
smb.ad.conf [for domain auth instance]
[global]
realm = dc.blah
security = ads
winbind separator = ^
winbind nested groups = yes
winbind enum groups = yes
winbind enum users = yes
password server = domaincont.dc.blah
# client code page = 437
interfaces = [ip removed].11/24
bind interfaces only = true
workgroup = dc
netbios name = smbserverAD
netbios aliases = smbserverAD
server string = smbserverAD
encrypt passwords = Yes
map to guest = Bad User
passwd program = /usr/bin/passwd
log file = /var/log/samba/log.ad.smbd
lock directory = /var/lib/samba.ad
time server = Yes
kernel oplocks = No
oplock break wait time = 20
deadtime = 10
keepalive = 30
admin users = %D^george
os level = 34
local master = Yes
preferred master = No
# domain master for WORKGROUP
domain master = Yes
# The only WINS server
wins support = Yes
# Local master browsers on the different subnets plus locale1 domain master
remote browse sync = [ips removed]
remote announce = [ip removed]/WORKGROUP [ip removed]/PHONE
unix extensions = Yes
log level = 2
max log size = 1048576
syslog = 0
printing = CUPS
printcap name = CUPS
use client driver = Yes
socket options = SO_KEEPALIVE IPTOS_LOWDELAY TCP_NODELAY
# veto files = /*.eml/*.nws/riched20.dll/*.{*}-
browseable = Yes
writeable = Yes
create mask = 0600
directory mask = 0700
locking = Yes
Here's a snippit of the logs from the panic.
Apr 15 08:00:11 shalimar smbd[28735]: [2011/04/15 08:00:11, 0]
printing/printing_db.c:get_print_db_byname(40)
Apr 15 08:00:11 shalimar smbd[28735]: PANIC: assert failed at
printing/printing_db.c(40)
Apr 15 08:00:11 shalimar smbd[28735]: [2011/04/15 08:00:11, 0]
lib/substitute.c:alloc_sub_basic(463)
Apr 15 08:00:11 shalimar smbd[28735]: alloc_sub_basic: NULL source
string! This should not happen
Apr 15 08:00:11 shalimar smbd[28735]: [2011/04/15 08:00:11, 0]
lib/fault.c:fault_report(41)
Apr 15 08:00:11 shalimar smbd[28735]:
==============================================================Apr 15 08:00:11
shalimar smbd[28735]: [2011/04/15 08:00:11, 0]
lib/fault.c:fault_report(42)
Apr 15 08:00:11 shalimar smbd[28735]: INTERNAL ERROR: Signal 11 in pid
28735 (3.0.36-7.1-2365-SUSE-CODE10)
Apr 15 08:00:11 shalimar smbd[28735]: Please read the Trouble-Shooting
section of the Samba3-HOWTO
Apr 15 08:00:11 shalimar smbd[28735]: [2011/04/15 08:00:11, 0]
lib/fault.c:fault_report(44)
Apr 15 08:00:11 shalimar smbd[28735]:
Apr 15 08:00:11 shalimar smbd[28735]: From:
http://www.samba.org/samba/docs/Samba3-HOWTO.pdf
Apr 15 08:00:11 shalimar smbd[28735]: [2011/04/15 08:00:11, 0]
lib/fault.c:fault_report(45)
Apr 15 08:00:11 shalimar smbd[28735]:
==============================================================Apr 15 08:00:11
shalimar smbd[28735]: [2011/04/15 08:00:11, 0]
lib/util.c:smb_panic(1633)
Apr 15 08:00:11 shalimar smbd[28735]: PANIC (pid 28735): internal error
Apr 15 08:00:19 shalimar smbd[28735]: BACKTRACE: 17 stack frames:
Apr 15 08:00:19 shalimar smbd[28735]: #0
/usr/sbin/smbd(log_stack_trace+0x1c) [0x555555773f8c]
Apr 15 08:00:19 shalimar smbd[28735]: #1
/usr/sbin/smbd(smb_panic+0x41) [0x555555774081]
Apr 15 08:00:19 shalimar smbd[28735]: #2 /usr/sbin/smbd [0x555555761ea2]
Apr 15 08:00:20 shalimar smbd[28735]: #3 /lib64/libc.so.6
[0x2b2fdfbadc10]
Apr 15 08:00:20 shalimar smbd[28735]: #4
/usr/sbin/smbd(Get_Pwnam_alloc+0x2b) [0x55555576693b]
Apr 15 08:00:20 shalimar smbd[28735]: #5
/usr/sbin/smbd(Get_Pwnam+0xb) [0x555555766c9b]
Apr 15 08:00:20 shalimar smbd[28735]: #6
/usr/sbin/smbd(get_user_home_dir+0x9) [0x555555766cd9]
Apr 15 08:00:20 shalimar smbd[28735]: #7 /usr/sbin/smbd [0x555555779fb4]
Apr 15 08:00:20 shalimar smbd[28735]: #8
/usr/sbin/smbd(standard_sub_advanced+0x26) [0x55555577a866]
Apr 15 08:00:20 shalimar smbd[28735]: #9 /usr/sbin/smbd [0x555555794d0e]
Apr 15 08:00:20 shalimar smbd[28735]: #10
/usr/sbin/smbd(print_queue_status+0x38a) [0x55555579547a]
Apr 15 08:00:20 shalimar smbd[28735]: #11
/usr/sbin/smbd(update_monitored_printq_cache+0x43) [0x5555556a1293]
Apr 15 08:00:20 shalimar smbd[28735]: #12 /usr/sbin/smbd [0x555555628a6b]
Apr 15 08:00:20 shalimar smbd[28735]: #13
/usr/sbin/smbd(smbd_process+0x430) [0x555555629640]
Apr 15 08:00:20 shalimar smbd[28735]: #14 /usr/sbin/smbd(main+0x1203)
[0x555555828903]
Apr 15 08:00:20 shalimar smbd[28735]: #15
/lib64/libc.so.6(__libc_start_main+0xf4) [0x2b2fdfb9b154]
Apr 15 08:00:20 shalimar smbd[28735]: #16 /usr/sbin/smbd [0x5555555bc779]
Apr 15 08:00:20 shalimar smbd[28735]: [2011/04/15 08:00:20, 0]
lib/fault.c:dump_core(181)
Apr 15 08:00:20 shalimar smbd[28735]: dumping core in
/var/log/samba/cores/smbd
Apr 15 08:00:20 shalimar smbd[28735]:
--
Matt Ingram
Intermediate Unix Administrator, IS
Canadian Bank Note Company, Limited
\m/
