visage

Hi, I currently have 4x3Tb drives in an array. Yesterday one if the data drives is showing as failed. I've bought 2x8Tb drives. So obviously one will become the new parity drive, and one will become a data drive. My NAS only has 4 bays so one of the good 3Tb drive will be removed. Is this a two stage process? Replace bad 3Tb drive with first new drive, rebuild array, then replace old parity drive with second new drive and then rebuild? Or is there a simpler way?

Thanks. One of the drives needs a tree rebuild apparently...

Looks like its not the USB stick. Bought a new one, clean install, started up, same symptom. I connected a monitor up to it after it went unresponsive, and it looks like there was a kernel panic. By that time the root cause had scrolled off the screen, but I did see part of the error mentioned reiserFS. So presumably I should restart the server, but not the array, then do a file system check?

I recently had an issue where I found my server was unresponsive. So I did a hard reset, and took a look - I saw lots of read only filesystem warnings, and looking in the plugins directory it looked like the file system was corrupted. SO I put my USB stick in my desktop and it reported it was corrupted. So I did a repair, put it back in the server. It booted, noted an unclean shutdown, then started aparity check. A minute or so later it went unresponsive again. SO I tried a completely clean install. Just copied the super.dat file and the key file. Same thing happen - system came up, started parity check and then became unresponsive. Interestingly each time I took the USB stick out to try a fix it was reported as corrupted. So im leaning towards a duff USB stick, where its OK enough to boot but once the array starts to come up it breaks, but before I go and get another I was wondering if anyone had any other things I could try?

My unraid box has worked fine for a while, but a couple of days ago we had a power cut. AFter strting up agin the core aray is fine, but any Docker containers I start up are causing the whole array to become unresponsive. The culprit always seems to be BTRFS errors, though even if I delete the docker.img file they always come back. Ive done a full party check, and dont see any errors, so why would a standalone img file have internal file-system errors?

Thanks for your reply. Looking at various wiki pages It seems that I should be able to just follow the unraid 6 clean install instructions, using the key file from my old installation, and then when prompted request a new key. In terms of restoring the array, ive got 4 data drives + parity - my understanding is that the parity drive will show as 'unformatted', so just assigning the others as data drives and that one as parity will see the array working (albeit probably needing a parity check...) Then i'll just need to install the unraid 6 verions of the apps I use. Does that sound right? Thanks!

... or is this natures way of telling me to just upgrade to v6?

So I bouned my server last night...and it wouldnt come back. So I connected it to a monitor to see what was happening during startup, and it just gets to a blinking cursor and thats it. Ive double checked all the BIOS options and its correctly set to boot from USB. SO I took the USB key and put it in a windows machine and it says the drive is corrupted....in disk manager its listed as a RAW format. So its broken, right? I have a backup from the 5th June - can I just get a new usb stick, follow the setup steps and then just copy the backup back to it?

The bit in my original; post is the only activity at the time I see the problem. That said, I think Ive fixed it. I created a new directory alongside the parent of the directory where I had the problem, and moved everything to it, then deleted the problematc one. And now it seems OK...

PFA syslog and smart reports. Nothing obviously wrong. Some errors on sde, but they seem to be historical, not recent. files.zip

So I recently had a pretty big failure. Two drives died, and I had to rebuild my array, having lost a fair bit of data. Now, the array is seemingly back to normal. All green lights, parity check run without error. But there are still intermittent erros when using the array; root@Tower:/mnt/user/MediaShare# mkdir TV mkdir: cannot create directory `TV': Input/output error The log is full of errors relating to md2: Jun 14 20:40:40 Tower shfs/user: shfs_mkdir: mkdir: /mnt/disk2/MediaShare/TV (5) Input/output error (Errors) Jun 14 20:40:40 Tower kernel: REISERFS warning (device md2): jdm-20001 reiserfs_xattr_get: Invalid magic for xattr (system.posix_acl_default) associated with [1953724787 1882090853 0x7869736f UNKNOWN] (Minor Issues) Ive run reiserfsck --check /dev/md2 and it completed without error. Not sure what to do next