Need for cleanup of idle winbindd child processes...

Volker Lendecke Volker.Lendecke at SerNet.DE
Tue Apr 22 02:57:59 MDT 2014


On Mon, Apr 21, 2014 at 01:32:10PM +0000, Ravindra Channabasappa wrote:
> Volker,
>   
>     In my setup there are only two winbindd processes. One
>     parent winbindd and the other one for the configured
>     domain.
> I have not enabled trusted domain configuration.  Do you
> still think mmap'ing tdb files has issues?  But I observe
> that for every run of 5000 users login, RSS size increases
> by ~500 KB (ps -eo output) and never comes down.

Yes, I still believe it can contribute to indicated memory
use. Have you tried the "use mmap = no" option? This would
eliminate that source of RSS use.

> I tried checking through 'smbcontrol --pool-usage'
> command. Output of the same is attached for before the
> load test and after the test.
> I could not figure much difference between the outputs.
> About valgrind output- having problem in getting valgrind
> run under jail environment where winbindd is running.

Ok. The pool-usage does not indicate anything that grows, so
it must be somewhere else. In the thread I missed the
information what base operating system you are running on.
In case it is Linux, the file /proc/<pid>/smaps shows which
memory segment uses how much memory. If possible, can you
compare those values before/after? Here you would also see
what happens to the tdb files, if they grow. If however it
is heap memory which is growing, there is no reasonable
other way than to run winbind under valgrind
--tool=memcheck, sorry for that.

> 
> Also, in another thread - "RE: [PATCH] Unix datagram
> socket messaging"- I see that patch mentions about cleanup
> of winbindd processes as well.

This cleanup relates to something different, not memory or
child processes. It relates to cleaning up unix domain
datagram sockets with a changed internal messaging scheme.
Processes should clean up behind themselves, the 15-minute
interval is deliberately chosen so long because it will only
clean up behind faulty or crashed processes. If that happens
very frequently, we have other problems to solve I guess.

> If so, IMO based on internal testing 15 minutes to cleanup
> child processes looks long. Is it fine to reduce the time
> to 5 minutes or make this configurable through smb.conf ?
> I hope we can apply this patch for Samba-4.1.3...

Sure, this will be configurable. But the specific 15 minute
cleanup interval in the thread is not relevant in 4.1.3,
which does not yet have the new messaging scheme.

With best regards,

Volker Lendecke

-- 
SerNet GmbH, Bahnhofsallee 1b, 37081 Göttingen
phone: +49-551-370000-0, fax: +49-551-370000-9
AG Göttingen, HRB 2816, GF: Dr. Johannes Loxen
http://www.sernet.de, mailto:kontakt at sernet.de


More information about the samba-technical mailing list