Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 8 Aug 2001 11:44:13 -0700
From:      "Brian O'Shea" <boshea@ricochet.net>
To:        Terry Lambert <tlambert2@mindspring.com>
Cc:        freebsd-hackers@FreeBSD.ORG
Subject:   Re: Tuning the 4.1-R kernel for networking
Message-ID:  <20010808114413.E529@ricochet.net>
In-Reply-To: <3B710F75.FF20B38@mindspring.com>; from tlambert2@mindspring.com on Wed, Aug 08, 2001 at 03:07:49AM -0700
References:  <20010807213320.D529@ricochet.net> <3B710F75.FF20B38@mindspring.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Wow, much useful feedback.  Thanks.

Terry, your general formula for nmbclusters per connection is pretty
much what I am looking for.  Great stuff.

> Frankly, it sounds like your application is bad; does it limit
> itself to 150 connections, or is it trying to make as many
> connections as it possible can make?  If so, then no matter
> how you tune the system, your program will always hit its head
> on a resource limit eventually.

The application attempts to create a specified number of client
connections (i.e. even in the absence of resource limits, it will only
create 150 client connections, in this particular configuration).

> > I compiled a debug kernel and panicked the system while it was in the
> > state described above, in case that is any use.  I don't know how to
> > analyze the crash dump to determine where the problem is.  Any
> > suggestions are welcome.
> 
> Did it panic from being in the state, or did you break to the
> debugger and force a crash?

I forced the crash to get a core file, in case it was useful in
debugging.

> You didn't _boot_ the kernel with the debugging symbols, right?

Uhh, ... :)
I booted the debug kernel (with debugging symbols) for one test so that
I could get a crash dump.  I have since booted only the stripped kernel.

Thanks to everyone for your help.
-brian

p.s.  Upgrading the machine is on the list of things to do.  I'm a
bit hesitant to keep up to date with -STABLE as I have had some bad
experiences doing that.  Possibly a monthly upgrade while closely
monitoring the -stable mailing list for known problems would be
feasible.

-- 
Brian O'Shea
<boshea@ricochet.net>

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010808114413.E529>