Fedora 27 RAID Resync on every boot, reopened [closed]

asked 2018-11-08 22:16:08 -0600

iliffe gravatar image

I had thought that the answer suggested before would resove this problem: echo idle > /sys/block/md127/md/sync_action but it doesn't work. Re-reading the sync_action file after issuing this command shows it to still be set to "resync" and my workstation is still unresponsive. mdadm --detail shows "state: clean, resyncing" and my workstaion is still unusable for hours. This behaviour arrived with Fedora update version 27-18.100 or the previous one, within the last three weeks.

There is very little data (a few tens of Gb on a 6 Tb RAID 5 system) and it seems to checking the entire enpty disk.

Any ideas?

Thanks in advance for your help.

edit retag flag offensive reopen merge delete

Closed for the following reason question is not relevant or outdated by iliffe
close date 2018-11-20 21:52:39.636188

Comments

A mdadm RAID will always resync the whole disks, despite the usage of the disks. Did you consider that this could be a hardware problem and did check your disks with smartctl -a? Also look into dmesg and /var/log/messages for problems. For me it also seems uncommon that the workstation is "unusabe" during the resync. It will put load on the disks and the computer, but a resync should not render the whole workstation useless.

thomaswood gravatar imagethomaswood ( 2018-11-10 11:30:25 -0600 )edit