Unfortunately no one can be told what fun_plug is - you have to see it for yourself.
You are not logged in.
I have a CH3SNAS with the lastest firmware version (1.5?) and funplug 0.5.
Funplugs runs the fancontroller script for some more silence. Last month I started using NTPD because the time was always wrong, what I didn't know what that this fucks with the temperature readings and the fanspeed. So after a month I noticed that the web interface said that the temp was 0c.
I think the harddrives (2x WD red 2TB in JBOD) got a little to warm, because I can't list certain directories trough SMB anymore (can list those trough SSH though...) and filetransfer trough SMB does about 250KB/s.
Logs say i need to run e2fsck, and that's where I get stuck. I can't unmount HD_a2, because it's always in use by something (SSHD or telnet).
I tried this solution, and I can at least run e2fsck, but mdadm can't properly assemble the JBOD array.
mdadm --detail /dev/md0
Shows a total size of only ~2TB, in stead of 4TB, so naturally e2fsck starts complaining that something is wrong. When I boot normally the array get assembled properly, so the array is not broken.
I ran
e2fsck -n /dev/md0
So it only checks, and it found some errors, but I can figure out how to get those fixed.
Offline
That mdadm story is very strange. I can imagine the mdadm in that package is not compatible with the (newer) version 1.x metadata. But in that case it should fail to assemble the array anyway. Unless you have had a (single disk) V0.9 array, and later created a V1.x dual disk array. In that case the old V0.9 header still might be readable.
Anyway, you can connect the disks to any Linux system to do the filesystem check.
Offline
But when I boot normally, and when I boot the 'e2fsck package' I see
Version : 0.90
So it seems the versions are matching. I never had a single disk in the NAS.
Problem with taking the disks out is that I am currently not in the same country as the NAS, so that's going to be difficult.
I saw in another topic that you can remount HD_a2 as read only and then run e2fsck, or that a recipe for disaster?
Offline
Dannisi wrote:
I saw in another topic that you can remount HD_a2 as read only and then run e2fsck, or that a recipe for disaster?
The only reason I can think of why you shouldn't fsck a mounted partition, is because the filesystem in changed under the cache, without notifying it. When the cache flushes disaster happens.
When a partition is mounted readonly, the cache will never flush, so that should be fine. But maybe the filesystem driver will crash, when it tries to access the partition, based on cached information, which is no longer valid.
AFAIK a filesystem driver which detects filesystem inconsistencies will just remount it read-only, to protect against futher damage. So that shouldn't hurt you.
You'll have to reboot the box after fsck finishes, as there is no way (AFAIK) to force the filesystem driver to use the new contents, other than un- and remounting it.
Offline