https://community.hpe.com/t5/disk-enclosures/msa-failure/m-p/5200203#M43401 solution was to make sure all firmware was up to date. The rather than having the 2 x msa's on a channel each, daisy chained them. The system then recognised them and re-established the mirrors. Once all had settled down, put them back onto they're own channels.<BR /><BR /> Tue, 08 Dec 2009 11:54:17 GMT halcyondays 2009-12-08T11:54:17Z https://community.hpe.com/t5/disk-enclosures/msa-failure/m-p/5200202#M43400 dl360g5 with p800 ctrl connected to 2 msa 50's on separate channels. 6 disks in each, forming mirrored pairs across the msa's ie, msa 1 disk 1 is mirrored to msa 2 disk 1. one of the msa's died. installed new back plane.<BR /><BR />mirror has not automatically re-established, all arrays in interim recovery.<BR /><BR />I believe this is because the repaired msa is classed as a new one (different box number reported in utils) not the old one.<BR /><BR />is it possible to re-number the repaired msa to give it the old box number (and therefore kicking in the mirrors recovery????) - or how do I remove the 'dead' disks from the mirrors and add the 'new' ones (although the disks are exactly the same, just a different address in the repaired msa)<BR /><BR />Oracle DB server so trying to have 0 down time.....<BR /><BR />Thanks ! ;)<BR /><BR />Jonty Tue, 22 Sep 2009 20:26:39 GMT https://community.hpe.com/t5/disk-enclosures/msa-failure/m-p/5200202#M43400 halcyondays 2009-09-22T20:26:39Z https://community.hpe.com/t5/disk-enclosures/msa-failure/m-p/5200203#M43401 solution was to make sure all firmware was up to date. The rather than having the 2 x msa's on a channel each, daisy chained them. The system then recognised them and re-established the mirrors. Once all had settled down, put them back onto they're own channels.<BR /><BR /> Tue, 08 Dec 2009 11:54:17 GMT https://community.hpe.com/t5/disk-enclosures/msa-failure/m-p/5200203#M43401 halcyondays 2009-12-08T11:54:17Z