Hi everyone,

Lately I've been noticing sporadic (ie: irregular patterns, so far
happened only twice) tdb corruption with Samba causing it to reject any
new connections. This is easily fixed by restarting the daemons, and or
flushing the /var/lock/samba/ files. But of course I have to be here to do
that, and it's ghastly when it happens during operating hours.

I can't figure out what's causing the problems because when they occur,
nothing special is happening or happened. The first time this happened was
around 5pm when load was normal. The second time was a little before 6am
when I was the only one here doing some maintenance on the workstations.

Here's a copy of the Samba log of one of the machines (similar for them
all). I wonder what's causing these errors. I'm running Samba 2.2.2-10
(Debian) on Linux kernel 2.4.17-xfs.

[2002/01/24 05:52:55, 0] tdb/tdbutil.c:tdb_log(342)
  tdb(/var/lock/samba/connections.tdb): rec_free_read bad magic 0x26011999 at 
offset=73728
[2002/01/24 05:52:55, 0] smbd/connection.c:claim_connection(188)
  claim_connection: tdb_store failed with error Corrupt database.
[2002/01/24 05:52:55, 1] lib/util_sock.c:get_socket_name(1003)
  Gethostbyaddr failed for 192.168.0.36
[2002/01/24 05:52:55, 0] tdb/tdbutil.c:tdb_log(342)
  tdb(/var/lock/samba/connections.tdb): rec_free_read bad magic 0x26011999 at 
offset=73728
[2002/01/24 05:52:55, 0] smbd/connection.c:claim_connection(188)
  claim_connection: tdb_store failed with error Corrupt database.
[2002/01/24 05:52:55, 1] smbd/service.c:make_connection(494)
  too many connections - rejected
[2002/01/24 05:52:55, 0] smbd/connection.c:yield_connection(62)
  yield_connection: tdb_delete for name  failed with error Record does not exist.
[2002/01/24 05:53:30, 0] tdb/tdbutil.c:tdb_log(342)
  tdb(/var/lock/samba/connections.tdb): rec_free_read bad magic 0x26011999 at 
offset=73728
[2002/01/24 05:53:30, 0] smbd/connection.c:claim_connection(188)
  claim_connection: tdb_store failed with error Corrupt database.
[2002/01/24 05:53:30, 1] lib/util_sock.c:get_socket_name(1003)
  Gethostbyaddr failed for 192.168.0.36
[2002/01/24 05:53:30, 0] tdb/tdbutil.c:tdb_log(342)
  tdb(/var/lock/samba/connections.tdb): rec_free_read bad magic 0x26011999 at 
offset=73728
[2002/01/24 05:53:30, 0] smbd/connection.c:claim_connection(188)
  claim_connection: tdb_store failed with error Corrupt database.
[2002/01/24 05:53:30, 1] smbd/service.c:make_connection(494)
  too many connections - rejected
[2002/01/24 05:53:31, 0] smbd/connection.c:yield_connection(62)
  yield_connection: tdb_delete for name  failed with error Record does not exist.


I do not know if this is relevant, but I've also been noticing something
to do with tdb problems in my syslog whenever users connect via dial-up.
They occur like so:

Jan 24 05:51:54 gusi pppd[11630]: pppd 2.4.1 started by root, uid 0
Jan 24 05:51:54 gusi pppd[11630]: tdb_store failed: Success
Jan 24 05:51:54 gusi pppd[11630]: tdb_store key failed: Corrupt database
Jan 24 05:51:54 gusi pppd[11630]: tdb_store failed: Corrupt database
Jan 24 05:51:54 gusi pppd[11630]: tdb_store key failed: Corrupt database
Jan 24 05:51:54 gusi pppd[11630]: tdb_store failed: Corrupt database
Jan 24 05:51:54 gusi pppd[11630]: tdb_store failed: Corrupt database
Jan 24 05:51:54 gusi pppd[11630]: Using interface ppp0
Jan 24 05:51:54 gusi pppd[11630]: tdb_store key failed: Corrupt database
Jan 24 05:51:54 gusi pppd[11630]: tdb_store failed: Corrupt database
Jan 24 05:51:54 gusi pppd[11630]: Connect: ppp0 <--> /dev/tts/0
Jan 24 05:51:54 gusi pppd[11630]: tdb_store failed: Corrupt database
Jan 24 05:51:54 gusi last message repeated 2 times
Jan 24 05:51:57 gusi pppd[11630]: user sstone logged in
Jan 24 05:51:57 gusi pppd[11630]: tdb_store failed: Corrupt database
Jan 24 05:51:57 gusi pppd[11630]: kernel does not support PPP filtering
Jan 24 05:51:58 gusi pppd[11630]: Deflate (15) compression enabled
Jan 24 05:51:58 gusi pppd[11630]: tdb_store failed: Corrupt database
Jan 24 05:51:58 gusi pppd[11630]: tdb_store key failed: Corrupt database
Jan 24 05:51:58 gusi pppd[11630]: tdb_store failed: Corrupt database
Jan 24 05:51:58 gusi pppd[11630]: found interface eth0 for proxy arp
Jan 24 05:51:58 gusi pppd[11630]: local  IP address 192.168.0.1
Jan 24 05:51:58 gusi pppd[11630]: remote IP address 192.168.0.90


Fortunately dial-up connections seem to be unaffected by these problems.
But still, this is definitely not good. Anyone with suggestions as to what
I should check?

And no, the filesystem seems to be intact, and all data (other than these
tdbs) seem to be intact, too. No reports of data loss or data corruption
from the users or from other programs.

 --> Jijo

--
Federico Sevilla III  :: [EMAIL PROTECTED]
Network Administrator :: The Leather Collection, Inc.
GnuPG Key: http://jijo.leathercollection.ph/jijo.gpg

_
Philippine Linux Users Group. Web site and archives at http://plug.linux.org.ph
To leave: send "unsubscribe" in the body to [EMAIL PROTECTED]

To subscribe to the Linux Newbies' List: send "subscribe" in the body to 
[EMAIL PROTECTED]

Reply via email to