From mboxrd@z Thu Jan 1 00:00:00 1970 From: Daniel Pittman Subject: Re: RAID5 temperorary loss two disk Date: Thu, 29 Apr 2004 12:46:08 +1000 Sender: linux-raid-owner@vger.kernel.org Message-ID: <87u0z3h573.fsf@enki.rimspace.net> References: <408D0FF0.5000608@uni-mainz.de> <20040428142849.68073.qmail@web14525.mail.yahoo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: To: linux-raid@vger.kernel.org List-Id: linux-raid.ids On Wed, 28 Apr 2004, Don Jessup wrote: > I have RAID5 volume that has two disk that have been marked as failed. > I believe this happen do to loss of communication with the disk. NOT > because of a total disk failure. I assume if I can get the superblocks > back into a good state I can reassemble the RAID5 and try to read what > data I can off the volume. Does anyone know how to force RAID5 > superblocks into a state so I can do reads from the volume? I really > need help on this. By the way the kernel version is 2.4.22. mdadm has a --force option to the assemble command which I have used to recover from a similar situation. Don't forget to fsck the volume afterwards, though, since it will have bad data on it. Daniel -- The Believer had better face himself and ask squarely: Do I literally believe "God" has a penis? If the answer is no, then it seems only logical to drop the ridiculous practice of referring to "God" as "he." -- Robert Anton Wilson