From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vladimir Bashkirtsev Subject: RBD image max simultaneous mounts Date: Mon, 17 Sep 2012 12:13:07 +0930 Message-ID: <50568E3B.3060403@bashkirtsev.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail.logics.net.au ([150.101.56.178]:46390 "EHLO mail.logics.net.au" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752940Ab2IQCoR (ORCPT ); Sun, 16 Sep 2012 22:44:17 -0400 Received: from x.logics.net.au (gw.logics.net.au [150.101.235.251] (may be forged)) (authenticated bits=0) by mail.logics.net.au (8.14.5/8.14.1) with ESMTP id q8H2h5Fl001640 (version=TLSv1/SSLv3 cipher=DHE-RSA-CAMELLIA256-SHA bits=256 verify=NO) for ; Mon, 17 Sep 2012 12:13:05 +0930 Sender: ceph-devel-owner@vger.kernel.org List-ID: To: ceph-devel Dear devs, We just had an incident where two instances of the same VM started up on different hosts and mounted the same RBD image. Image had ext4 partition and it became corrupted quickly. Therefore it would be good to have some setting in RBD info like max_simultaneous_mounts which will dictate how many simultaneous mounts of the same RBD image may be in existence. Obviously such glusterfs would not have such issue in RBD image but old trusty extfs fails miserably. It may be hard to implement in ceph due to distributed nature. Then basic check in QEMU-RBD driver would be good as well. Regards, Vladimir