This is a potentially dumb question... I had a 5 disk Raid5 with 3TB disks. One disk threw some errors (but still seemed to work), and i decided to replace it. I actually bought two drives, thinking I'd keep one as a cold spare, but when i got to replacing the drive (after days of load testing to see if I'd gotten a dud), I figured maybe I'd use the spare as a hot spare, or make a Raid6 array. In the end I decided on a Raid6, as it is the recommended way to go these days for large disks. When I did the add for the first disk, it started rebuilding the raid5, and said it'd only take like 5 hours, then I added the second disk and it skyrocketed to about 116 hours (7000s or so). I presume this is expected, due to having to re-stripe all of the data, and re-calculate the parity at the same time. How much faster would this have finished if I had just waited for the first drive to finish rebuilding and done the reshape as a separate step? I decided to just let this reshape continue, its about 30% the way through, and this is actually the backup array for my main NAS, so it isn't actually a problem if it dies mid-reshape. I'm just curious about the alternatives to how I did it. I know i could have just copied the old disk to one of the new ones, and skipped the entire rebuild on one disk, but i didn't feel like it ;D -- Thomas Fjellstrom thomas@xxxxxxxxxxxxx -- 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
