David Wright <lily...@lionunicorn.co.uk> writes: > On Tue 29 Nov 2016 at 09:37:21 (+0100), David Kastrup wrote: >> David Wright <lily...@lionunicorn.co.uk> writes: >> >> > On Mon 28 Nov 2016 at 21:26:17 (+0000), Karlin High wrote: >> >> On 11/28/2016 2:12 PM, David Wright wrote: >> >> > So it should be worth booting from a live linux CD to mount the >> >> > partitions to check their contents, and to reinstall Grub >> >> > (or whatever you use to boot) into the MBR. >> >> >> >> The thing to do IMMEDIATELY is make a "drive image backup." >> > >> > That would certainly be the action to take if the drive was giving >> > disk errors. >> >> It's also the action to take if you are dealing with damage to the data >> structures. > > That doesn't necessarily buy you any advantage in the case you > outlined. There are risks in making bit for bit copies of a drive. > For starters, you're _writing_ to a device, whereas attempting to > mount the partitions readonly involves _no_ writing to any device. > > Only on the 18th, I read a post where a user was trying to make an > image of a drive, and was relying on the order they plugged in the > two drives to get the kernel to assign the "correct" /dev/sdX values > to the two drives so that they could then follow some remotely > posted instructions for making the copy. Talk about tail wagging dog!
Just as a P.S.: am in the mirroring stage now (my father acquired a new somewhat larger SSD disk with some mSATA interface like his computer uses internally (the kind of crap that exists these days...) and a suitable USB adapter). And indeed the externally attached drive (via USB) was named /dev/sda while the internal drive the Linux rescue environment booted from was /dev/sdb. So this time round the root disk system not being able to mount stuff on /dev/sda* had more than one reason. But I was surprised that a USB-connected drive (even though connected at bootup time) was listed before the internal mSATA one. Without checking via fdisk prior to starting the copy I would indeed have assumed the reverse. -- David Kastrup _______________________________________________ lilypond-user mailing list lilypond-user@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-user