From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wido den Hollander Subject: Re: Port no For data Transfer between OSDs Date: Wed, 05 Sep 2012 15:04:46 +0200 Message-ID: <50474DEE.6060304@widodh.nl> References: <50471810.6000608@widodh.nl> <50473CA2.3050806@widodh.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from smtp02.mail.pcextreme.nl ([109.72.87.138]:60654 "EHLO smtp02.mail.pcextreme.nl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750981Ab2IENEt (ORCPT ); Wed, 5 Sep 2012 09:04:49 -0400 In-Reply-To: <50473CA2.3050806@widodh.nl> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: hemant surale Cc: ceph-devel@vger.kernel.org On 09/05/2012 01:50 PM, Wido den Hollander wrote: > On 09/05/2012 01:20 PM, hemant surale wrote: >> Hi Wido , Community , >> >> if we use number =2 >> in cmd as follows >> "ceph osd pool set size <#no>" >> >> does it means for perticular Object we have 2 replica Or total 2 >> copies in cluster? >> > > That will set replication to "2". That means, every object stored in the > cluster will be stored twice. > Sorry, I have to correct myself there. Every object in that (RADOS) POOL is stored twice, not the cluster. This is a pool specific setting, not cluster-wide. Wido > It is NOT object + 2, no it's object * 2. > > Wido > >> >> Regards, >> Hemant Surale. >> >> On Wed, Sep 5, 2012 at 2:44 PM, Wido den Hollander >> wrote: >>> On 09/05/2012 07:30 AM, hemant surale wrote: >>>> >>>> Hi Community , >>>> I want to trace the data transfer rates occurring between >>>> osds while injecting data into ceph cluster. Can you please tell me >>>> what ports should I monitor or keep track of? >>>> Thanks ! >>> >>> >>> OSDs start binding from port 6800 and counting up from there. >>> >>> For example, a machine with 4 OSDs running: >>> >>> root@atom0:~# netstat -nap|grep LISTEN|grep -E ^tcp|grep ceph-osd >>> tcp6 0 0 :::6800 :::* LISTEN >>> 24329/ceph-osd >>> tcp6 0 0 :::6801 :::* LISTEN >>> 24329/ceph-osd >>> tcp6 0 0 :::6802 :::* LISTEN >>> 24329/ceph-osd >>> tcp6 0 0 :::6803 :::* LISTEN >>> 24596/ceph-osd >>> tcp6 0 0 :::6804 :::* LISTEN >>> 24596/ceph-osd >>> tcp6 0 0 :::6805 :::* LISTEN >>> 24596/ceph-osd >>> tcp6 0 0 :::6806 :::* LISTEN >>> 25316/ceph-osd >>> tcp6 0 0 :::6807 :::* LISTEN >>> 25316/ceph-osd >>> tcp6 0 0 :::6808 :::* LISTEN >>> 25316/ceph-osd >>> tcp6 0 0 :::6809 :::* LISTEN >>> 25839/ceph-osd >>> tcp6 0 0 :::6810 :::* LISTEN >>> 25839/ceph-osd >>> tcp6 0 0 :::6811 :::* LISTEN >>> 25839/ceph-osd >>> root@atom0:~# >>> >>> Wido >>> >>>> >>>> Regards, >>>> Hemant Surale. >>>> -- >>>> To unsubscribe from this list: send the line "unsubscribe >>>> ceph-devel" in >>>> the body of a message to majordomo@vger.kernel.org >>>> More majordomo info at http://vger.kernel.org/majordomo-info.html >>>> >>> >> -- >> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in >> the body of a message to majordomo@vger.kernel.org >> More majordomo info at http://vger.kernel.org/majordomo-info.html >> >