samba - Mar 2005 - corruption in the locking tdb, samba panics

Hello,

last friday I switched my users from an old server running 2.2.12 to a 
new one running 3.0.10 (that I've been testing myself and with 
smbtorture). After a while I had to switch back to the old server 
because some users were having serious problems. This is one of the logs 
(the others are very similar):

[2005/03/04 12:25:05, 0] tdb/tdbutil.c:tdb_log(725)
   tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 80 ltype=1 
(Resource deadlock avoided)
[2005/03/04 12:25:05, 0] smbd/oplock.c:remove_oplock(272)
   remove_oplock: failed to lock share entry for file eplan4/N/0408/Se.gif
[2005/03/04 12:25:05, 0] smbd/reply.c:reply_lockingX(4559)
   reply_lockingX: error in removing oplock on file eplan4/N/0408/Se.gif
[2005/03/04 12:25:05, 0] tdb/tdbutil.c:tdb_log(725)
   tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 80 ltype=1 
(Resource deadlock avoided)
[2005/03/04 12:25:05, 0] tdb/tdbutil.c:tdb_log(725)
   tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 80 ltype=0 
(Resource deadlock avoided)
[2005/03/04 12:25:05, 0] tdb/tdbutil.c:tdb_log(725)
   tdb(/var/cache/samba/locking.tdb): tdb_unlock: count is 0
[2005/03/04 12:25:05, 0] smbd/oplock.c:request_oplock_break(977)
   request_oplock_break: PANIC : breaking our own oplock requested for 
dev = fe00, inode = 745870013, file_id = 1471 a
nd no fsp found !
[2005/03/04 12:25:05, 0] lib/util.c:smb_panic2(1482)
   PANIC: request_oplock_break: no fsp found for our own oplock

[2005/03/04 12:25:05, 0] lib/util.c:smb_panic2(1490)
   BACKTRACE: 14 stack frames:
    #0 smbd(smb_panic2+0x1b6) [0x81cdf1a]
    #1 smbd(smb_panic+0x19) [0x81cdd62]
    #2 smbd(request_oplock_break+0x1d5) [0x81fca07]
    #3 smbd [0x80bf613]
    #4 smbd(open_file_shared1+0x801) [0x80c08ce]
    #5 smbd(open_file_shared+0x51) [0x80c00cb]
    #6 smbd(reply_open_and_X+0x2bc) [0x80a14c4]
    #7 smbd [0x80d04c4]
    #8 smbd [0x80d0574]
    #9 smbd(process_smb+0x1eb) [0x80d08b9]
    #10 smbd(smbd_process+0x170) [0x80d149c]
    #11 smbd(main+0x7d4) [0x8240f8c]
    #12 /lib/tls/libc.so.6(__libc_start_main+0xed) [0x402ec95d]
    #13 smbd(fsetxattr+0x31) [0x8078aa1]

I've looked at the changelog for 3.0.11 but I see nothing there 
regarding this problem. I also searched google and I found something 
similar here:

http://tinyurl.com/3nhdj

but, alas, no followup or solution.
The new server is running mandrake linux 10.0, samba comes from the 
updates rpms (3.0.10-0.1.100mdk) and it's running kernel 2.4.25 (I 
cannot use 2.6 since this is a proliant server --not my decision-- and 
its closed source hardware monitoring module only works with 2.4).
The server is configured to use ldap for authentication 
(openldap-2.1.25-6mdk) if it matters.

Bye
-- 
Luca Olivetti
Wetron Automatizaci?n S.A. http://www.wetron.es/
Tel. +34 93 5883004      Fax +34 93 5883007

Luca Olivetti wrote:
> I've looked at the changelog for 3.0.11 but I see nothing there 
> regarding this problem. I also searched google and I found something 
> similar here:
> 
> http://tinyurl.com/3nhdj
> 
> but, alas, no followup or solution.
> The new server is running mandrake linux 10.0, samba comes from the 
> updates rpms (3.0.10-0.1.100mdk) and it's running kernel 2.4.25 (I 
> cannot use 2.6 since this is a proliant server --not my decision-- and 
> its closed source hardware monitoring module only works with 2.4).
> The server is configured to use ldap for authentication 
> (openldap-2.1.25-6mdk) if it matters.

I forgot to mention that this happened with win98, windows 2000 and 
xp+sp2 clients.

Bye
-- 
Luca Olivetti
Wetron Automatizaci?n S.A. http://www.wetron.es/
Tel. +34 93 5883004      Fax +34 93 5883007

Luca Olivetti wrote:
> Hello,
> 
> last friday I switched my users from an old server running 2.2.12 to a 
> new one running 3.0.10 (that I've been testing myself and with 
> smbtorture). After a while I had to switch back to the old server 
> because some users were having serious problems. This is one of the logs 
> (the others are very similar):
> 
> [2005/03/04 12:25:05, 0] tdb/tdbutil.c:tdb_log(725)
>   tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 80 ltype=1 
> (Resource deadlock avoided)
> [2005/03/04 12:25:05, 0] smbd/oplock.c:remove_oplock(272)
>   remove_oplock: failed to lock share entry for file eplan4/N/0408/Se.gif
> [2005/03/04 12:25:05, 0] smbd/reply.c:reply_lockingX(4559)
>   reply_lockingX: error in removing oplock on file eplan4/N/0408/Se.gif
> [2005/03/04 12:25:05, 0] tdb/tdbutil.c:tdb_log(725)
>   tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 80 ltype=1 
> (Resource deadlock avoided)
> [2005/03/04 12:25:05, 0] tdb/tdbutil.c:tdb_log(725)
>   tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 80 ltype=0 
> (Resource deadlock avoided)
> [2005/03/04 12:25:05, 0] tdb/tdbutil.c:tdb_log(725)
>   tdb(/var/cache/samba/locking.tdb): tdb_unlock: count is 0
> [2005/03/04 12:25:05, 0] smbd/oplock.c:request_oplock_break(977)
>   request_oplock_break: PANIC : breaking our own oplock requested for 
> dev = fe00, inode = 745870013, file_id = 1471 a
> nd no fsp found !
> [2005/03/04 12:25:05, 0] lib/util.c:smb_panic2(1482)
>   PANIC: request_oplock_break: no fsp found for our own oplock
> 
> [2005/03/04 12:25:05, 0] lib/util.c:smb_panic2(1490)
>   BACKTRACE: 14 stack frames:
>    #0 smbd(smb_panic2+0x1b6) [0x81cdf1a]
>    #1 smbd(smb_panic+0x19) [0x81cdd62]
>    #2 smbd(request_oplock_break+0x1d5) [0x81fca07]
>    #3 smbd [0x80bf613]
>    #4 smbd(open_file_shared1+0x801) [0x80c08ce]
>    #5 smbd(open_file_shared+0x51) [0x80c00cb]
>    #6 smbd(reply_open_and_X+0x2bc) [0x80a14c4]
>    #7 smbd [0x80d04c4]
>    #8 smbd [0x80d0574]
>    #9 smbd(process_smb+0x1eb) [0x80d08b9]
>    #10 smbd(smbd_process+0x170) [0x80d149c]
>    #11 smbd(main+0x7d4) [0x8240f8c]
>    #12 /lib/tls/libc.so.6(__libc_start_main+0xed) [0x402ec95d]
>    #13 smbd(fsetxattr+0x31) [0x8078aa1]
> 
> I've looked at the changelog for 3.0.11 but I see nothing there 
> regarding this problem. I also searched google and I found something 
> similar here:
> 
> http://tinyurl.com/3nhdj
As a followup, I have switched to the new server the problematic share 
only, and it has been working no problem for the last two days.
I don't know what cured my issues: I upgraded samba from 3.0.10 to 
3.0.11, changed kernel from 2.4.25-8mdk to 2.6.3-25mdk (now hp proliant 
diagnostics are working even under 2.6, but I don't remember any hp 
advisory telling that), upgraded to smbldap-tools 0.8.7 and re-imported 
all users' credentials in ldap (though I doubt that these two steps had 
anything to do with the solution of my problems).
I'm keeping my fingers crossed but it seems everything is working as 
expected.

Bye
-- 
Luca Olivetti
Wetron Automatizaci?n S.A. http://www.wetron.es/
Tel. +34 93 5883004      Fax +34 93 5883007