On Wed, Mar 24, 2010 at 3:10 PM, Brad Campbell <brad@xxxxxxxxxxx> wrote: > G'day all, > > Having just dropped a drive (it turns out 1TB SATA drives don't bounce) that > belonged in my 10 drive RAID-6 and having replaced it this afternoon, I had > a thought. > > When things go wrong (particularly those nasty things that cause the sick > feeling in the pit of your stomach) the best help is always available give > the right information. > > If there was a script that one could run when things were healthy that > dumped all the relevant information for safe keeping, would it reduce the > nightmare that is recovering a broken raid? > > I'd imagine an mdadm --examine on each component of a RAID, partition tables > and perhaps a --detail on each md device. Is there anything else that might > be handy that could be dumped to a verbose output just in case? > > Is there already such a thing? > > Regards, > Brad In my paranoia, I did this type of thing manually the other day. (Great minds think alike?) - I took the output of "lshw". Handy overview. Shows drive types, serial numbers, drive-to-controller connections. - It'd be very useful to have the UUID serials of each drive plus the exact order they are assembled into the raid. - And well, having a binary copy of all the superblocks and partition tables sure can't hurt. -- Kristleifur -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
