[SRU] PANIC (pid 18442): internal error in winbindd

Can you install samba-dbg and reproduce this please?

Thanks
chuck

Binary package hint: samba

Description: Ubuntu 10.04.2 LTS
Release: 10.04

I'm having a similar crash after samba and kernel update last night.
Winbind crashed 2 times today.

samba-dbg was not installed.

core dump mentioned in a log file

I have a similar problem. winbindd does start, but has crashed several times since installing 2:3.4.7~dfsg-1ubuntu3.5. Here is my gdb backtrace from core dump:

Core was generated by `/usr/sbin/winbindd'.
Program terminated with signal 6, Aborted.
#0 0x00007f4fbe8fba75 in raise () from /lib/libc.so.6
(gdb) bt
#0 0x00007f4fbe8fba75 in raise () from /lib/libc.so.6
#1 0x00007f4fbe8ff5c0 in abort () from /lib/libc.so.6
#2 0x00007f4fc0da1450 in dump_core () at lib/fault.c:337
#3 0x00007f4fc0db227e in smb_panic (why=<value optimized out>)
    at lib/util.c:1496
#4 0x00007f4fc0da184d in fault_report (sig=11) at lib/fault.c:52
#5 sig_fault (sig=11) at lib/fault.c:75
#6 <signal handler called>
#7 rpc_pipe_np_smb_conn (p=0x0) at rpc_client/rpc_transport_np.c:402
#8 0x00007f4fc0e8d048 in rpccli_set_timeout (rpc_cli=0x0, timeout=35000)
    at rpc_client/cli_pipe.c:2956
#9 0x00007f4fc0d277a1 in winbindd_lookup_sids (mem_ctx=0x7f4fc2ae8560,
    domain=<value optimized out>, num_sids=<value optimized out>,
    sids=<value optimized out>, domains=0x7fff185eaf90, names=0x7fff185eafa0,
    types=0x7fff185eaf98) at winbindd/winbindd_rpc.c:1221
#10 0x00007f4fc0d2a092 in lookup_groupmem (domain=0x7f4fc2b0a240,
    mem_ctx=0x7f4fc2ae8780, group_sid=0x7f4fc2b2e360,
    num_names=0x7fff185eb2dc, sid_mem=<value optimized out>,
    names=<value optimized out>, name_types=0x7fff185eb2a8)
    at winbindd/winbindd_ads.c:1124
#11 0x00007f4fc0d11503 in lookup_groupmem (domain=<value optimized out>,
    mem_ctx=<value optimized out>, group_sid=<value optimized out>,
    num_names=0x7fff185eb2dc, sid_mem=0x7fff185eb298, names=0x7fff185eb2a0,
---Type <return> to continue, or q <return> to quit---
    name_types=0x7fff185eb2a8) at winbindd/winbindd_cache.c:2211
#12 0x00007f4fc0d045e7 in expand_groups (domain=0x7f4fc2b0a240,
    state=<value optimized out>, group_sid=<value optimized out>,
    group_name_type=SID_NAME_DOM_GRP, num_gr_mem=<value optimized out>,
    gr_mem=<value optimized out>, gr_mem_len=0x7fff185eb668)
    at winbindd/winbindd_group.c:468
#13 fill_grent_mem (domain=0x7f4fc2b0a240, state=<value optimized out>,
    group_sid=<value optimized out>, group_name_type=SID_NAME_DOM_GRP,
    num_gr_mem=<value optimized out>, gr_mem=<value optimized out>,
    gr_mem_len=0x7fff185eb668) at winbindd/winbindd_group.c:604
#14 0x00007f4fc0d05a9f in getgrsid_sid2gid_recv (private_data=0x7f4fc2aa6890,
    success=<value optimized out>, gid=16777729)
    at winbindd/winbindd_group.c:866
#15 0x00007f4fc0cfeba0 in process_loop (argc=<value optimized out>,
    argv=<value optimized out>, envp=<value optimized out>)
    at winbindd/winbindd.c:1115
#16 main (argc=<value optimized out>, argv=<value optimized out>,
    envp=<value optimized out>) at winbindd/winbindd.c:1440

Same problem happen to me
ubuntu version: 10.04.2 x86_64
winbind version: 2:3.4.7~dfsg-1ubuntu3.5
And have the similar thread debugging message:

[Thread debugging using libthread_db enabled]
0x00007fb594f32f7e in waitpid () from /lib/libc.so.6
#0 0x00007fb594f32f7e in waitpid () from /lib/libc.so.6
#1 0x00007fb594eca7e9 in ?? () from /lib/libc.so.6
#2 0x00007fb597373264 in smb_panic (why=<value optimized out>)
    at lib/util.c:1486
#3 0x00007fb59736284d in fault_report (sig=11) at lib/fault.c:52
#4 sig_fault (sig=11) at lib/fault.c:75
#5 <signal handler called>
#6 rpc_pipe_np_smb_conn (p=0x0) at rpc_client/rpc_transport_np.c:402
#7 0x00007fb59744e048 in rpccli_set_timeout (rpc_cli=0x0, timeout=35000)
    at rpc_client/cli_pipe.c:2956
#8 0x00007fb5972e87a1 in winbindd_lookup_sids (mem_ctx=0x7fb597fb2b50,
    domain=<value optimized out>, num_sids=<value optimized out>,
    sids=<value optimized out>, domains=0x7fff2af6f1c0, names=0x7fff2af6f1d0,
    types=0x7fff2af6f1c8) at winbindd/winbindd_rpc.c:1221
#9 0x00007fb5972eb092 in lookup_groupmem (domain=0x7fb598041950,
    mem_ctx=0x7fb597fe8950, group_sid=0x7fb598084a90,
    num_names=0x7fff2af6f50c, sid_mem=<value optimized out>,
    names=<value optimized out>, name_types=0x7fff2af6f4d8)
    at winbindd/winbindd_ads.c:1124
#10 0x00007fb5972d2503 in lookup_groupmem (domain=<value optimized out>,
    mem_ctx=<value optimized out>, group_sid=<value optimized out>,
    num_names=0x7fff2af6f50c, sid_mem=0x7fff2af6f4c8, names=0x7fff2af6f4d0,
    name_types=0x7fff2af6f4d8) at winbindd/winbindd_cache.c:2211
#11 0x00007fb5972c55e7 in expand_groups (domain=0x7fb598041950,
    state=<value optimized out>, group_sid=<value optimized out>,
    group_name_type=SID_NAME_DOM_GRP, num_gr_mem=<value optimized out>,
    gr_mem=<value optimized out>, gr_mem_len=0x7fff2af6f898)
    at winbindd/winbindd_group.c:468
#12 fill_grent_mem (domain=0x7fb598041950, state=<value optimized out>,
    group_sid=<value optimized out>, group_name_type=SID_NAME_DOM_GRP,
    num_gr_mem=<value optimized out>, gr_mem=<value optimized out>,
    gr_mem_len=0x7fff2af6f898) at winbindd/winbindd_group.c:604
#13 0x00007fb5972c6a9f in getgrsid_sid2gid_recv (private_data=0x7fb597fd0200,
    success=<value optimized out>, gid=11610) at winbindd/winbindd_group.c:866
#14 0x00007fb5972bfba0 in process_loop (argc=<value optimized out>,
    argv=<value optimized out>, envp=<value optimized out>)
    at winbindd/winbindd.c:1115
#15 main (argc=<value optimized out>, argv=<value optimized out>,
    envp=<value optimized out>) at winbindd/winbindd.c:1440
A debugging session is active.

This is fixed in natty.

Regards
chuck

Will release the fix in lucid?

I agree with sense. I am experiencing this on LTS servers with over 30 thousand users that can't login whenever this gets triggered.

I also have this problem with lucid.
winbindd crashes every 2 days after my last upgrade (~1 week ago)

I have the same problem with lucid. Downgrading to 2:3.4.7~dfsg-1ubuntu3.4 solved the problem but this is no satisfying solution. I think a bugfix for lucid LTS is strictly necessary!

Statement of Impact: The last SRU update for fixing bug #623342 introduced a regression that was fixed upstream. I have backported from the samba git tree. I have attached the debdiff which fixes this bug.

How to reproduce:

1. Install the lucid samba update
2. Restart winbindd

APPROVED: version 2:3.4.7~dfsg-1ubuntu3.6 uploaded to lucid-proposed should be accepted.

Accepted samba into lucid-proposed, the package will build now and be available in a few hours. Please test and give feedback here. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you in advance!

Tested samba 2:3.4.7~dfsg-1ubuntu3.6 for 2 days, no panic. Thanks.

Am 04/14/2011 07:07 PM, schrieb Martin Pitt:
> Accepted samba into lucid-proposed, the package will build now and be
> available in a few hours. Please test and give feedback here. See
> https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to
> enable and use -proposed. Thank you in advance!
Installed version from lucid-proposed yesterday - no problems yet, seems
fixed.

cu
Michael

It's been 36 hours, and have not got the "rpc_pipe_np_smb_conn" crash, as I call it, on any of the 3 machines that I installed the 2:3.4.7~dfsg-1ubuntu3.6 package.
Will still be monitoring this, and may even install it on a few more machines, today.

It's been 6 days, and the 5 machines that I have this patch on now, are doing fine.

Ich bin bis 28.04.2011 abwesend.

Ich werde Ihre Nachricht nach meiner Rückkehr beantworten. In dringenden
Fällen wenden Sie sich bitte an meinen Kollegen Helmut Juchems:
<email address hidden>

Hinweis: Dies ist eine automatische Antwort auf Ihre Nachricht "[Bug
736757] Re: [SRU] PANIC (pid 18442): internal error in winbindd" gesendet
am 25.04.2011 19:32:40.

Diese ist die einzige Benachrichtigung, die Sie empfangen werden, während
diese Person abwesend ist.

I pushed out the patch to 24 more machines ( 29 total) on Monday 2011-04-25 6:30 PM. All is fine since then.
Pushing out the patch to the rest of the 80 machines, right now.

Everything still working fine on all 80 machines since 2011-04-27.

Just chiming in to say that 3.4.7~dfsg-1ubuntu3.6 from lucid-proposed also fixed this problem on our servers as well.

We've gone from daily winbind panics to no crashes since. We haven't noticed any side-effects either.

Any idea if or when this makes it into lucid-updates?

After applying the patches, winbind is behaving nicely and we also have no more panics!

Please escalate this into lucid-updates!

To all who would like to see this released into lucid-updates, we need to have some verification of bug #558793 as well, since both of them are included in the package currently in -proposed.

> To all who would like to see this released into lucid-updates, we need
> to have some verification of bug #558793 as well, since both of them are
> included in the package currently in -proposed.

We just migrated our AD Servers to 2008 this weekend, and had the problems
described in #558793 (according to my colleagues who did the migration).

Also, one Linux server that was still running the release version of
10.04.02 had intermittent problems authenticating users (but no winbind
crashes).

Both problems went away after upgrading to the -proposed-upgrades versions
of samba+winbind.

cu
Michael Schwingen

Thanks Michael. I'll consider that a verification, and have marked the bug as such.

This bug was fixed in the package samba - 2:3.4.7~dfsg-1ubuntu3.6

---------------
samba (2:3.4.7~dfsg-1ubuntu3.6) lucid-proposed; urgency=low

  * debian/patches/backport-lp558793.patch: Fixes net ads dns register
    failing with Windows 2008 R2 domain. Backported from
    v3-4-stable tree (aaf6e34115c2bece85d0dc4f11951b38e9d77637
    and 0e1aec461ed69c82987cf86e6b1beb6862cc0fb4). (LP: #558793)
  * debian/patches/backport-lp736757.patch: Fix regression introduced
    by previous SRU bug fix. Backported from the v3-4-stable tree
    (995e7e500327e662b7ef2b37c83c92e75f2360bf) (LP: #736757)
 -- Chuck Short <email address hidden> Wed, 13 Apr 2011 11:13:44 -0400