From mboxrd@z Thu Jan 1 00:00:00 1970 From: Piotr =?utf-8?B?RGHFgmVr?= Subject: Re: Questions about profiling ceph Date: Tue, 22 Nov 2016 21:00:37 +0100 Message-ID: <20161122200037.GD26997@predictor> References: Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Return-path: Received: from predictor.org.pl ([185.5.97.54]:53201 "EHLO predictor.org.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755807AbcKVT6Y (ORCPT ); Tue, 22 Nov 2016 14:58:24 -0500 Content-Disposition: inline In-Reply-To: Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Victor Denisov Cc: ceph-devel On Tue, Nov 22, 2016 at 11:42:48AM -0800, Victor Denisov wrote: > Hello ceph developers, > > I'm learning how to profile ceph and came across this page: > http://docs.ceph.com/docs/giant/rados/troubleshooting/cpu-profiling/ > > The page suggests to use oprofile. I compiled oprofile version > 1.1.0(http://oprofile.sourceforge.net/faq/), but it doesn't have > opcontrol binary. > > Am I doing something wrong? Is it an outdated procedure for profiling? > Is there a newer one? Well, they actually noted (in the faq you linked) that opcontrol is a legacy tool. And the cpu profiling guide is for "Giant" Ceph release (also quite legacy). Anyway, operf + opreport tandem is sufficient for me and I'm sure it will be enough for you. -- Piotr Dałek branch@predictor.org.pl http://blog.predictor.org.pl