[no subject]

sekishi.zefyris.com>
From: Francois Tigeot <ftigeot@wolfpond.org>
Subject: Re: Machine unresponsive with cache_lock: blocked on... message
Date: Mon, 22 Mar 2010 12:02:25 +0100
BestServHost: crater.dragonflybsd.org
List-Post: <mailto:bugs@crater.dragonflybsd.org>
List-Subscribe: <mailto:bugs-request@crater.dragonflybsd.org?body=subscribe>
List-Unsubscribe: <mailto:bugs-request@crater.dragonflybsd.org?body=unsubscribe>
List-Help: <mailto:bugs-request@crater.dragonflybsd.org?body=help>
List-Owner: <mailto:owner-bugs@crater.dragonflybsd.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
References: <20100320083423.GA992@sekishi.zefyris.com> <201003201834.o2KIYJ4h004349@apollo.backplane.com> <20100320184452.GA1507@sekishi.zefyris.com> <1240.220.233.36.106.1269126548.squirrel@webmail.exetel.com.au> <201003202324.o2KNOrlX006114@apollo.backplane.com> <1366.220.233.36.106.1269129077.squirrel@webmail.exetel.com.au> <201003211824.o2LIOZUs015643@apollo.backplane.com> <20100321190708.GA63881@sekishi.zefyris.com> <201003211914.o2LJEBNo016061@apollo.backplane.com> <20100322083907.GA1801@
sekishi.zefyris.com>
In-Reply-To: <20100322083907.GA1801@sekishi.zefyris.com>
Sender: bugs-errors@crater.dragonflybsd.org
Errors-To: bugs-errors@crater.dragonflybsd.org
Lines: 19
NNTP-Posting-Host: 216.240.41.25
X-Trace: 1269255929 crater_reader.dragonflybsd.org 55376 216.240.41.25
Xref: crater_reader.dragonflybsd.org dragonfly.bugs:11559

On Mon, Mar 22, 2010 at 09:39:07AM +0100, Francois Tigeot wrote:
> On Sun, Mar 21, 2010 at 12:14:11PM -0700, Matthew Dillon wrote:
> > 
> >     I'd like to know if this solves your particular problem (the paging
> >     issue, not the crashing issue), and if so I will change the default
> >     cycle point for the release.  So play with it.
> 
> With vfs.vm_cycle_point = 40 my server was much more responsive this morning.
> Almost no programs had to be paged in when first accessed.
> 
> I have now decreased vm_cycle_point to 32.

With 32 the situation is slightly worse: some processes are once again paged
out when not constantly accessed.

I'm currently testing with vfs.vm_cycle_point = 36.

-- 
Francois Tigeot