Mailing List Archive
tlug.jp Mailing List tlug archive tlug Mailing List Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]Re: [tlug] How to proceed with RAID drive failure
- Date: Wed, 13 Jul 2011 15:03:11 +0200
- From: Attila Kinali <attila@example.com>
- Subject: Re: [tlug] How to proceed with RAID drive failure
- References: <CA+kCxRZLF78pq0C41vH2tT5KfPndzwxxPJQs-KF-FMs9iN2tkw@example.com> <87hb6st4bs.fsf@example.com> <CA+kCxRaMoCfeuWm6fA+dDUpsd0TxF1H6t-mL41GuqdO7U5ucmA@example.com> <CAFv52OC5HLxUTs1dgqXa=RSCYfGMH4Jktu1Uei0Py82yQ8vh1g@example.com> <CA+kCxRb-qROMGDp6c=sihoW-R-8W7-6ydVkniKJFcU299r+z9w@example.com>
- Organization: GEHIRN
On Wed, 13 Jul 2011 21:08:14 +0900 Martin G <ebisumartin@example.com> wrote: > As you suggested, I unplugged one drive at a time, rebooted, and the one > that didn't work was the one I replaced. > > I managed to get a drive of almost exactly the same specs. It's plugged in > now, and according to the partition manager, it's all unallocated space so > far. I was kind of hoping that there might be some kind of automagic > detection and setup, but it doesn't look like that's going to happen. > > It's a software RAID that was put together from the Ubuntu alternative > install CD when I first installed Ubuntu on this machine, so I'm assuming > there's some configuration tool somewhere that I can use to get the new > drive to mirror the old drive, but beats me where the heck that is. That's the one of the two big disadvantages of software RAID, there is no automagic reconstruction, because the RAID system does not know whether a drive will end up in the RAID or whether it will be used for something else. If you have a RAID1, then both disks were most probably set up the same way, ie you can copy the partition table over to the new disk and rebuild the RAID. You can use sfdisk -d /dev/source | sfdisk /dev/dest to copy the partition table over or, if you are using GPT sgdisk -R /dev/dest /dev/source # copy partition table, be carefull about order sgdisk -G /dev/dest # randomize the GUIDs an then it's a mdadm --add /dev/mdX /dev/destY where X and Y are the "partition" numbers you were using. Also check whether you had any non-RAID partitions (like for swap), those have to be initialized on their own (mkswap). While the RAID is recovering from the failure, you can watch its progress at /proc/mdstat HTH Attila Kinali -- The trouble with you, Shev, is you don't say anything until you've saved up a whole truckload of damned heavy brick arguments and then you dump them all out and never look at the bleeding body mangled beneath the heap -- Tirin, The Dispossessed, U. Le Guin
- Follow-Ups:
- Re: [tlug] How to proceed with RAID drive failure
- From: Josh Glover
- References:
- [tlug] How to proceed with RAID drive failure
- From: Martin G
- [tlug] How to proceed with RAID drive failure
- From: Stephen J. Turnbull
- Re: [tlug] How to proceed with RAID drive failure
- From: Martin G
- Re: [tlug] How to proceed with RAID drive failure
- From: Josh Glover
- Re: [tlug] How to proceed with RAID drive failure
- From: Martin G
Home | Main Index | Thread Index
- Prev by Date: Re: [tlug] How to proceed with RAID drive failure
- Next by Date: Re: [tlug] How to proceed with RAID drive failure
- Previous by thread: Re: [tlug] How to proceed with RAID drive failure
- Next by thread: Re: [tlug] How to proceed with RAID drive failure
- Index(es):
Home Page Mailing List Linux and Japan TLUG Members Links