From mboxrd@z Thu Jan 1 00:00:00 1970 From: Adam Kwolek Subject: [PATCH 0/4] Raid0 metadata cannot be recognized Date: Tue, 15 Mar 2011 14:48:16 +0100 Message-ID: <20110315134000.32643.1818.stgit@gklab-128-013.igk.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Return-path: Sender: linux-raid-owner@vger.kernel.org To: neilb@suse.de Cc: linux-raid@vger.kernel.org, dan.j.williams@intel.com, ed.ciechanowski@intel.com, wojciech.neubauer@intel.com List-Id: linux-raid.ids Hi, I've executed unit tests on today's devel3-2 and I've found that expansion cannot be executed due to fact that imsm metadata cannot be found after creation. Looking patches I've found that problem was introduced by patch: Manage/external: for external metadata, add_to_super needs lock on container. It is due to fact that metadata is not written by monitor for raid0. Problem is fixed by patch: FIX: Write metadata when mdmon is not running Other patches addresses problems/i.e. inconsistency in closing handle/ that I've noticed during investigation. BR Adam --- Adam Kwolek (4): FIX: ping_monitor() usage causes memory leaks FIX: Handle has to be closed FIX: Write metadata when mdmon is not running FIX: Ping monitor when mdmon is running only Assemble.c | 2 +- Create.c | 2 +- Grow.c | 2 +- Incremental.c | 10 ++++------ Manage.c | 10 +++++++++- Monitor.c | 2 +- msg.c | 14 ++++++++++++++ msg.h | 1 + 8 files changed, 32 insertions(+), 11 deletions(-) -- Signature