[Samba] hanging smbd(s) revisited

[Gerald (Jerry) Carter]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

William Jojo wrote:

> filemon summary shows fd 3 with 250MB/s for a process 
> that's not running any more and IBM cannot tell me why that
> happens. (we're  AIX 5.2 TL-08-1 and 3.0.21b) So that's why
> I haven't been that suspicious of Samba.

I wonder if this is an fcntl wake up bug like the only
that was previously found on Solaris.

Jeremy/Tridge, I can't remember which file it was in
samba.org/~tridge/junkcode.  Can you help me out?

> If Jerry or Jeremy can make a suggestion of what to look for, 
> I'll set it up on my end as well. We don't have Samba
> actually *stop*, it's just *reeeeealy* slow and context
> switches go fromabout 8-20k/sec(with 120k
> syscalls) to over 60k/sec (with only 30k syscalls).

This sounds really similar to me.  It was a kernel bug
on solaris when waking up processes waiting on a byte
range lock.

> The other noticeable trait is the time it takes to 
> stop smbd. If I didn't know any better, I'd swear they are
> trying to unravel some seriously nested stack of calls or
> complex pieces of data. It takes almost two minutes for
> all the smbd's to clear out after stopping from SWAT.

This is normal.  smbd handles signals in band so the we
receive the TERM signal and then schedule the shutdown
when we get done with what we are doing.





cheers, jerry
=====================================================================
I live in a Reply-to-All world                -----------------------
Samba                                    ------- http://www.samba.org
Centeris                         -----------  http://www.centeris.com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFEAIIgIR7qMdg1EfYRAroAAKCxxhU8+LDt3wl9lHkoIEdm7NUluwCglU+l
jF57iwY68h+WAQTJq064iRQ=
=7ApC
-----END PGP SIGNATURE-----