Multiple smbd processes generated

[Varnavas, Frank]

I've seen this happen when one or more NFS mounts hang.  The processes in
question hang on the NFS access and cant be killed. The NT redirector times
the session out, closes the connection, and opens a new one. This will
happen every 45 seconds (the default redirector timeout). 

Limiting your samba exports to local filesystems should prevent this, as
well as improve response time.

Frank Varnavas

> 
> From:   William Jojo <jojowil at hvcc.edu>
> To:     A.Boswell at uea.ac.uk
> Subject: Re: Multiple smbd processes generated
> 
> I've seen this on AIX 4.3.2 with 2.0.6. We chalked it up to 
> Win2k because we had
> all NT4 and one NT4 box upgraded to Lose2k which was part of 
> the domain at that
> point. We had to reboot on two consecutive Tuesdays (no 
> relevance really) and
> we've been fine ever since...perhaps we picked the wrong culprit?
> 
> We had 156 blocked processes waiting for a resource - what I 
> have no idea
> because the log was barren. We had 1342 processes that were 
> smbd's. And get this
> - LOTS of idle - NOT wait time.
> 
> Bill
> 
> Andrew Boswell wrote:
> > 
> > We have an occasional problem which manifests with multiple 
> processes
> > being created for a particular user.  For example, for a 
> user "xy004":
> > 
> > xy004       8463  0.0  0.0 6.02M   0K ??       IW   
> 11:48:59     0:02.02 smbd
> > xy004       9426  0.0  0.0 5.99M   0K ??       IW   
> 11:52:18     0:03.68 smbd
> > xy004      10433  0.0  0.0 5.81M   0K ??       IW   
> 12:17:20     0:00.85 smbd
> > xy004      12211  0.0  0.0 5.81M   0K ??       IW   
> 12:29:50     0:00.88 smbd
> > xy004      12588  0.0  0.0 5.81M   0K ??       IW   
> 12:16:05     0:00.55 smbd
> > xy004      12968  0.0  0.0 5.81M   0K ??       IW   
> 12:18:40     0:00.85 smbd
> > xy004      13069  0.0  0.0 5.81M   0K ??       IW   
> 12:19:55     0:00.86 smbd
> > xy004      13223  0.0  0.0 5.81M   0K ??       IW   
> 12:21:10     0:00.83 smbd
> > xy004      13396  0.0  0.0 5.81M   0K ??       IW   
> 12:22:27     0:00.86 smbd
> > xy004      13579  0.0  0.0 5.81M   0K ??       IW   
> 12:23:42     0:00.82 smbd
> > xy004      13772  0.0  0.0 5.81M   0K ??       IW   
> 12:24:52     0:00.87 smbd
> > xy004      13887  0.0  0.0 5.81M   0K ??       IW   
> 12:26:07     0:00.83 smbd
> > xy004      14064  0.0  0.0 5.81M   0K ??       IW   
> 12:27:18     0:00.82 smbd
> > xy004      14220  0.0  0.0 5.81M   0K ??       IW   
> 12:28:35     0:00.85 smbd
> > xy004      14743  0.0  0.0 5.81M   0K ??       IW   
> 12:31:05     0:00.98 smbd
> > xy004      14963  0.0  0.0 5.81M   0K ??       IW   
> 12:32:30     0:00.93 smbd
> > xy004      15609  0.0  0.0 5.81M   0K ??       IW   
> 12:33:55     0:00.87 smbd
> > 
> > This has manifested in two ways:
> > 
> > 1) For just a single user.  The processes remain following
> > disconnection from Samba, and *CANNOT* be killed (with 
> "kill -9" etc).
> > In this situtaion they have status <defunct>, and the user cannot
> > reconnect to Samba.  This has caused a problem to the user - they
> > couldn't connect to Samba, except if the Samba server was 
> rebooted.  As
> > we have parallel servers, we have however been able to tell users to
> > use a different server.
> > 
> > 2) For multiple users.  This has happened once, but required an
> > emergency reboot of the server due to the process table limit being
> > reached - the above process list was obtained just before 
> shutting down
> > the server in this case.
> > 
> > We are using Compaq Tru64 Unix (ex- Digital Unix) and Samba 2.0.6.
> > 
> > Has anyone else encountered these problems and know what 
> causes them?
> > I can find no evidence in the log files of anything wrong 
> for the user
> > to which this happens.
> > 
> > TIA
> > 
> > Andrew
> > 
> > 
> ======================================================================
> > Dr Andrew Boswell               email : A.Boswell at uea..ac.uk
> > School Liaison Consultant       phone : +44-1603-593856
> > IT and Computing Services       fax   : +44-1603-593467
> > University of East Anglia
> > Norwich, NR4 7TJ, UK
> 
> -- 
> 
> 
>         /------------------------------------------------------\
>         |                                                      |
>         |                 William E. Jojo, Jr.                 |
>         |                                                      |
>         |         Senior Systems and Network Specialist        |
>         |                                                      |
>         |            Hudson Valley Community College           |
>         |                                                      |
>         |                    (518) 629 7540                    |
>         |                                                      |
>         |                   jojowil at hvcc.edu                   |
>         |                                                      |
>         \------------------------------------------------------/
>