From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stefan Priebe Subject: Re: FS / Kernel question choosing the correct kernel version Date: Tue, 26 Jun 2012 18:15:02 +0200 Message-ID: <4FE9E006.6030404@profihost.ag> References: <4FE60A65.2030800@profihost.ag> <4FE97BE8.6010603@profihost.ag> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail.profihost.ag ([85.158.179.208]:52655 "EHLO mail.profihost.ag" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758580Ab2FZQPG (ORCPT ); Tue, 26 Jun 2012 12:15:06 -0400 In-Reply-To: <4FE97BE8.6010603@profihost.ag> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Sage Weil Cc: "ceph-devel@vger.kernel.org" , "hch@infradead.org >> Christoph Hellwig" Am 26.06.2012 11:07, schrieb Stefan Priebe: > Am 26.06.2012 00:11, schrieb Sage Weil: >> On Sat, 23 Jun 2012, Stefan Priebe wrote: >>> Hi, >>> >>> i got stuck while selecting the right FS for ceph / RBD. >>> >>> XFS: >>> - deadlock / hung task under 3.0.34 in xfs_ilock / xfs_buf_lock while >>> syncfs >> >> There was an ilock fix that went into 3.4, IIRC. Have you tried vanilla >> 3.4? We are seeing some lockdep noise currently, but no deadlocks yet. > > Will test 3.4.4 soon. With btrfs i have 13-14.000 iops while doing random 4k writes from kvm. With XFS and 3.4.4 i have 814 ;-( and got a lot of slow request noise in ceph.log ;-( so 814 iops are not an option to me. Or is there anything special i've to respect while using XFS. Greets Stefan