From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stefan Priebe Subject: Re: session lost, hunting for new mon Date: Sat, 19 May 2012 10:13:30 +0200 Message-ID: <4FB7562A.5080103@profihost.ag> References: <4FB72E28.2040709@profihost.ag> <4FB738F3.1060602@inktank.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail.profihost.ag ([85.158.179.208]:60938 "EHLO mail.profihost.ag" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752126Ab2ESINa (ORCPT ); Sat, 19 May 2012 04:13:30 -0400 In-Reply-To: <4FB738F3.1060602@inktank.com> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Josh Durgin Cc: ceph-devel@vger.kernel.org Am 19.05.2012 08:08, schrieb Josh Durgin: > On 05/18/2012 10:22 PM, Stefan Priebe wrote: >> Hi, >> >> while doing my first tests with ceph i'm always struggling about these >> kernel messages: >> >> [ 273.332472] libceph: client0 fsid d3aa8d88-3aa8-4f74-bb87-aa8de4975a84 >> [ 273.351025] libceph: mon1 10.0.255.101:6789 session established >> [ 459.830296] libceph: mon1 10.0.255.101:6789 socket closed >> [ 459.847044] libceph: mon1 10.0.255.101:6789 session lost, hunting for >> new mon >> [ 472.226073] libceph: mon0 10.0.255.100:6789 session established >> >> Always while stress testing the session to the mon's got lost. Any ideas? > > These aren't a problem - it's just the connection to a monitor being > lost, and the client connecting to another one. But what is causing this? I'm also seeing: [ 623.554698] ceph: mds0 caps stale [ 650.190504] ceph: mds0 caps renewed My biggest problem is that random reads and writes are awfully slow. Stefan