From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from 254.102.124.202.static.netspeed.net.nz ([202.124.102.254]:45715 "EHLO mx.anathoth.net" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1751402AbbDPBlk (ORCPT ); Wed, 15 Apr 2015 21:41:40 -0400 Received: from localhost (localhost [127.0.0.1]) by mx.anathoth.net (Postfix) with ESMTP id 6FB5841C60 for ; Thu, 16 Apr 2015 13:41:35 +1200 (NZST) Received: from [49.224.132.98] (98.132.224.49.dyn.cust.vf.net.nz [49.224.132.98]) (Authenticated sender: matt@mattgrant.net.nz) by mx.anathoth.net (Postfix) with ESMTPSA for ; Thu, 16 Apr 2015 13:41:29 +1200 (NZST) Date: Thu, 16 Apr 2015 13:41:21 +1200 Subject: Btrfs-cleaner FS DoS issues From: Matt Grant To: linux-btrfs@vger.kernel.org MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Message-Id: <20150416014135.6FB5841C60@mx.anathoth.net> Sender: linux-btrfs-owner@vger.kernel.org List-ID: Hi! When ever I delete a large snapshot this stalls all the processes on my system for 30 minutes plus, kernel v19.2. Btrfs-cleaner is taking 100% CPU when completely stalled. Every few minutes its say 99.8% and satall abates and  processes/io happen. About to lodge a kernel bug about this as it is serious issue. Could some one look at making the clean up process more sensitive to when the system is idle? MD Raid is very good at this, and it should be possible to set this up. Best Regards, Matt Grant{.n++%ݶw{.n+{k~^nrzh&zzޗ++zfh~iz_j:+v)ߣm