Thanks for the responses. Richard: it takes that long because of the "-c -c" option. As Berend says, it runs badblocks which "beats up a block at a time" writing/reading patterns... looking for errors. The man page for mke2fs is helpful. Berend: THANK YOU FOR YOUR RESPONSE! I of course, didn't think of that. All I can say is WOW! Two weeks. I used to do this as a matter of course on ALL new drives but 2TB and above became prohibitively lengthy. I tried it on a 4TB drive IIRC and something like 85 hours comes to mind. That might have been a 2TB drive though. Life is too short. Sigh. As for using USB, I have several docking stations in use on a laptop, USB3.0... I have had drives "go away" and "come back" as other drive letters IF i make them too busy. I have written several bug reports on this, mostly on bugzilla.redhat.com. Things are getting better stability wise but didn't get much response from anyone. Even tried kernel.org... NO RESPONSE there. Sigh. I think it's the kernel reacting to prolonged delays in command execution... the system "thinks" the drive is gone due to a timeout of some kind and goes into error recovery... At least this is what it looks like. It is pure conjecture on my part though. Like the guy who complained to the Dr., "My arm hurts when I do that" to which the Dr. says, "Then don't do that". I don't make the drives as busy... doesn't hurt as much. Sigh... Again, THANKS to all who responded. George... On Sat, 2019-10-19 at 10:29 +0000, George R Goffe via users wrote: > Hi, > > I have a 2TB HDD with a possibly major problem. I have cleared the > data from it and have tried to run "mke2fs -t ext4 -j /dev/sdc1" but > after 24 hours it had just finished the 0xaa phase and started > reading when the drive seemingly dropped ready and was re-assigned as > a new device to /dev/sdf. Argh. I did get an ended message with an > ending block number. I'd like to restart at that point in time and > continue the media check. I don't see anything in the mke2fs man page > though. > > Has anyone seen or heard of a way to do this? mkfs -c just runs badblocks with the appropriate parameters. You can specify how many blocks to skip, for example badblocks /dev/sdc1 1234. You'll need to specify the block size correctly to badblocks, and you'll want badblocks -o /tmp/blocklist to save the bad blocks so mkfs -l /tmp/blocklist reads it back. To get the block size, format it without -c and look at the output or use dumpe2fs. Doing this over a USB2 cable took me two weeks last time I tried, for a 4TB drive. I hope it's faster for you. (related note: I find normal 2 and 4TB USB drives highly unreliable.) _______________________________________________ users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@xxxxxxxxxxxxxxxxxxxxxxx
