I have recently finished upgrading my workstation to include an existing file server systems disks, 6 sata drives-1TB each, in addition to two new disks on a sata card, 500GB each. I have managed to get the raid6 partition, approx 3.8GB working size, from the file server to be recognised and mounted ok. I have also deleted the original raid setup partitions from the file server that held the boot, root, *swap, home, and tempdata, so I now have approx 100GB free at the beginning of each hard disk, so 400GB workable storage if its added into the last remaining raid 6. I have superblock/metadata 1.2 and the drives are all GPT partitioned. Now my questions are... 1, Is it better/preferable to unmount the file system on the RAID? 2, Am I right in thinking that the the RAID must be running while I do work on it, I cant just tear down the array, delete and recreate the partition table entry to include the unused space on the 6 disks and then restart the array? 3, Assuming 2 is "it must be running" then is the following the best way to resize the array? 3a, mdadm /dev/md6 --fail /dev/sdX6 --remove /dev/sdX6 3b, delete and recreate the partition on /dev/sdX to use all the space 3c, mdadm --add /dev/md6 /dev/sdX 3d, let it resync then repeat for the other 5 drives. 3e, mdadm --grow /dev/md6 --size=max 3f, e2fsck -f /dev/md6 3g, resize2fs /dev/md6 3h, e2fsck -f /dev/md6 And hopefully the jobs done! 4, Are there any other things I should specify or do such as using an internal bitmap, or stopping it; or using a backupfile for 3e? 5, As the data on the partition does not get changed, only the partition start sector, is there a way to tell mdadm not to resync as in theory there has been no change to the contents of the drive after it was removed and added back, especially as the file system was unmounted at the time? Thanks in advance. -- 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
