runaway processes in 3.0a20

[Bradley W. Langhorst]

On Wed, 2002-12-11 at 16:42, Esh, Andrew wrote:

> Maybe have a look at /proc/17206 (or whatever PID is at the top of
> top), and see what is in the fd directory. 
the only unusual thing in there is
/var/cache/samba/printing/.tdb

here's the whole list:
rwx------    1 root     root           64 Dec 13 09:28 0 -> /dev/null
lrwx------    1 root     root           64 Dec 13 09:28 1 -> /dev/null
lrwx------    1 root     root           64 Dec 13 09:28 10 ->
/var/run/samba/namecache.tdb
lrwx------    1 root     root           64 Dec 13 09:28 11 ->
/var/run/samba/brlock.tdb
lr-x------    1 root     root           64 Dec 13 09:28 12 ->
pipe:[22422712]
l-wx------    1 root     root           64 Dec 13 09:28 13 ->
pipe:[22422712]
lrwx------    1 root     root           64 Dec 13 09:28 14 ->
socket:[22423024]
lrwx------    1 root     root           64 Dec 13 09:28 15 ->
/var/run/samba/locking.tdb
lrwx------    1 root     root           64 Dec 13 09:28 16 ->
/var/cache/samba/printing/printers.tdb
lrwx------    1 root     root           64 Dec 13 09:28 17 ->
/var/spool/samba/smbprn.00000200.I6wwkb
lrwx------    1 root     root           64 Dec 13 09:28 18 ->
/var/cache/samba/printing/lab_nec_unix.tdb
lrwx------    1 root     root           64 Dec 13 09:28 19 ->
/var/cache/samba/printing/lab_phaser.tdb
l-wx------    1 root     root           64 Dec 13 09:28 2 ->
/var/log/samba/log.smbd.jon kingsbury
lrwx------    1 root     root           64 Dec 13 09:28 20 ->
/var/cache/samba/printing/Lexmark212.tdb
lrwx------    1 root     root           64 Dec 13 09:28 21 ->
/var/lib/samba/ntdrivers.tdb
lrwx------    1 root     root           64 Dec 13 09:28 22 ->
/var/lib/samba/ntprinters.tdb
lrwx------    1 root     root           64 Dec 13 09:28 23 ->
/var/lib/samba/ntforms.tdb
lrwx------    1 root     root           64 Dec 13 09:28 24 ->
/var/lib/samba/share_info.tdb
lrwx------    1 root     root           64 Dec 13 09:28 25 ->
/var/lib/samba/registry.tdb
lrwx------    1 root     root           64 Dec 13 09:28 26 ->
/var/lib/samba/netlogon_unigrp.tdb
l-wx------    1 root     root           64 Dec 13 09:28 27 ->
/var/log/samba/log.smbd.hostname
lrwx------    1 root     root           64 Dec 13 09:28 28 ->
socket:[22423025]
lr-x------    1 root     root           64 Dec 13 09:28 29 ->
pipe:[22423026]
l-wx------    1 root     root           64 Dec 13 09:28 3 ->
pipe:[22422228]
l-wx------    1 root     root           64 Dec 13 09:28 30 ->
pipe:[22423026]
lrwx------    1 root     root           64 Dec 13 09:28 31 ->
socket:[22423064]
lrwx------    1 root     root           64 Dec 13 09:28 32 ->
/var/lib/samba/group_mapping.tdb
lrwx------    1 root     root           64 Dec 13 09:28 33 ->
/var/run/samba/sessionid.tdb
lrwx------    1 root     root           64 Dec 13 09:28 35 ->
/var/cache/samba/printing/.tdb
lr-x------    1 root     root           64 Dec 13 09:28 4 ->
/dev/urandom
lrwx------    1 root     root           64 Dec 13 09:28 5 ->
/var/lib/samba/secrets.tdb
lrwx------    1 root     root           64 Dec 13 09:28 6 ->
/var/lib/samba/account_policy.tdb
l-wx------    1 root     root           64 Dec 13 09:28 7 ->
/var/run/samba/smbd.pid
lrwx------    1 root     root           64 Dec 13 09:28 8 ->
/var/run/samba/messages.tdb
lrwx------    1 root     root           64 Dec 13 09:28 9 ->
/var/run/samba/connections.tdb


> If you have debug symbols, you could also attach with gdb and do a
> stack backtrace, just to see where it is. Also single step a little to
> find out where the loop is being closed, and on what condition.
here's the backtrace
(gdb) where
#0  0x08173b28 in rap_to_pjobid ()
#1  0x081750f4 in print_job_exists ()
#2  0x0807e638 in reply_trans ()
#3  0x080810e0 in api_reply ()
#4  0x080790c3 in send_trans_reply ()
#5  0x08079958 in reply_trans ()
#6  0x080a5b0e in respond_to_all_remaining_local_messages ()
#7  0x080a5bae in respond_to_all_remaining_local_messages ()
#8  0x080a5df3 in process_smb ()
#9  0x080a66c7 in smbd_process ()
#10 0x0807044f in main ()
#11 0x400e814f in __libc_start_main () from /lib/libc.so.6

I found a panic in the log for this one...

[2002/12/12 17:11:38, 2] lib/access.c:check_access(327)
  Allowed connection from  (132.177.45.114)
[2002/12/12 17:11:38, 0] lib/fault.c:fault_report(36)
  ===============================================================
[2002/12/12 17:11:38, 0] lib/fault.c:fault_report(37)
  INTERNAL ERROR: Signal 11 in pid 20669 (2.999+3.0.alpha20-4 for
Debian)
  Please read the file BUGS.txt in the distribution
[2002/12/12 17:11:38, 0] lib/fault.c:fault_report(39)
  ===============================================================
[2002/12/12 17:11:38, 0] lib/util.c:smb_panic(1151)
  PANIC: internal error
[2002/12/12 17:11:38, 2] passdb/pdb_ldap.c:ldapsam_open_connection(267)
  ldap_open_connection: connection opened

looks like something in the printing area?


brad

-- 
Bradley W. Langhorst <brad at langhorst.com>