On Mon, 2007-10-01 at 10:06 -0400, Mike Davis wrote:> After upgrading to 3.0.26a and moving to linux my member
> server gets hundreds of IPC$ connections when I run
> smbstatus. I also see in my logs the following...
>
> [2007/10/01 10:01:15, 0]
> lib/util_tdb.c:tdb_chainlock_with_timeout_internal(84)
> tdb_chainlock_with_timeout_internal: alarm (10) timed out
> for key VALDEZ in tdb /usr/local/samba/private/secrets.tdb
>
> I did a dump of secrets on my old server and there wasnt a
> key for Valdez there.
>
> Now valdez is of teh DC's for the Domain. I dont start
> seeing all of thsi until we get high activity. All the
> clients do is login into teh domain and then we have a bat
> file on the DC that mounts their home directory on this
> server.
>
> Can anyone point me in the right direction for tracking down
> why this is happening.
I guess you are describing a situation where you have a lot of process
hanging where the smbd is basically stuck.
How many smbd process do you have a around when this happen?
(ps xa |grep smbd|wc -l)
How many clients?
On what OS are you running this?
Is /usr/local/samba a local file system (ext2/3?)?
And most importantly can you reproduce this at will ?
Simo.
--
Simo Sorce
Samba Team GPL Compliance Officer
email: idra@samba.org
samba.org