From mboxrd@z Thu Jan 1 00:00:00 1970 From: KAMEZAWA Hiroyuki Subject: Re: [PATCH 9/9] ext3: do not throttle metadata and journal IO Date: Thu, 23 Apr 2009 14:37:30 +0900 Message-ID: <20090423143730.a289f0c1.kamezawa.hiroyu__14922.3254913979$1240465350$gmane$org@jp.fujitsu.com> References: <20090421174620.GD15541@mit.edu> <20090421181429.GO19637@balbir.in.ibm.com> <20090421191401.GF15541@mit.edu> <20090421204905.GA5573@linux> <20090422093349.1ee9ae82.kamezawa.hiroyu@jp.fujitsu.com> <20090422102153.9aec17b9.kamezawa.hiroyu@jp.fujitsu.com> <20090422102239.GA1935@linux> <20090423090535.ec419269.kamezawa.hiroyu@jp.fujitsu.com> <20090423012254.GZ15541@mit.edu> <20090423115419.c493266a.kamezawa.hiroyu@jp.fujitsu.com> <20090423043547.GB2723@mit.edu> <20090422215825.f83e1b27.akpm@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20090422215825.f83e1b27.akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org Errors-To: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org To: Andrew Morton Cc: randy.dunlap-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org, Theodore Tso , ngupta-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org, subrata-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org, Jens-FOgKQjlUJ6BQetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, agk-9JcytcrH/bA+uJoB2kUjGw@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Carl Henrik Lunde , dave-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org, fernando-gVGce1chcLdL9jVzuh4AOg@public.gmane.org, roberto-5KDOxZqKugI@public.gmane.org, containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, Axboe , matt-cT2on/YLNlBWk0Htik3J/w@public.gmane.org, dradford-cT2on/YLNlBWk0Htik3J/w@public.gmane.org, Paul Menage , Balbir Singh , Gui-FOgKQjlUJ6BQetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, eric.rannaud-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org, Andrea Righi List-Id: containers.vger.kernel.org On Wed, 22 Apr 2009 21:58:25 -0700 Andrew Morton wrote: > > Or at least that *someone* is > > doing so and can perhaps document how all of these knobs interact. > > After all, if they are going to be separate, and someone turns the I/O > > throttling knob without bothering to turn the write throttling knob > > --- what's going to happen? An OOM? That's not going to be very safe > > or friendly for the sysadmin who plans to be configuring the system. > > ... this problem. > Considering that low-io-limit cgroup as very slow device, the problem itself is not far from that the current kernel has. If per-bdi-dirty-ratio works well, we can write per-cgroup-dirty-ratio, I think. (I'll do if I find time.) But yes, configuration-how-to should be documented, finally. I hope sysadmins will not use some acrobatic configuration ;) Thanks, -Kame