Guys I am experiencing some trouble with unraid and I was hoping to get some advice before I screw this up

 

I noticed one of my drives was showing errors and I couldn't read something off of it - it was causing sagetv to lock up.

 

Anyway I rebooted my laptop (obviously unrelated) but I came back to the web interface and now it won't come back up.  I was able to access unmenu until a few minutes ago now it seems to crap out.  I have telnet access.  So not really sure or know what to do from here.  I do have the syslog in my browser session but I couldn't download it.  Attached is everything i copied and pasted.  Please provide any ideas here.  good news is I have a precleared drive just waiting but this is a 2TB drive that is like 1.8TB full so its going to take a while to rebuild I am sure.

 

This is 4.6 btw.

 

Thanks!

 

Neil

syslog.txt

It looks like disk6 is crapping out for some reason. The non-stop read errors are probably causing your unRAID server to run out of RAM and drop processes (e.g. emhttp, which would kill the web UI).

 

If you haven't yet, run a SMART report on disk6.

 

 

Well I couldn't really do anything and I lost network so I had to control alt delete the box.  It rebooted and is doing a parity test.  It was going well before I went to bed - said it was going to take about 6 hours.  Well I just checked it - excited it would be done like Xmas morning all over but I got a lump of coal in my stocking  

 

Total size: 1,953,514,552 KB

Current position: 741,624,588 (37.9%)

Estimated speed: 98 KB/sec

Estimated finish: 204674.7 minutes

 

 

So I have like another 10 years before it is done.  Im afraid if it craps out I lose parity?  How does that work?  I would be fine to just pull the disk and repair - I mean I am in a bad way here right now.  This isn't even reasonably quick.  That drive shows lots of errors so it is clearly the problem.  How does one go about killing the curent parity check then replace the drive and rebuild the raid?

 

Thanks,

 

Neil

 

ok I stopped the parity check - shut down - pulled the suspect drive and rebuilding on a new drive now.

 

My parity was old.  It takes so long to check that I just can't do it too often but I think I should schedule it monthly...  Any thoughts on this?

ok I stopped the parity check - shut down - pulled the suspect drive and rebuilding on a new drive now.

 

My parity was old.  It takes so long to check that I just can't do it too often but I think I should schedule it monthly...  Any thoughts on this?

Correct parity is never "old".

 

The primary benefit of regular parity checks is that it exercises the system, and should help you find a silently failing drive sooner rather than later. When a drive fails, it requires every other drive in the system to be perfectly read in order to reconstruct the failed drive smoothly. If you almost never spin up and read from some of your drives, whether because they are nearly empty, or rarely used, it's very possible the drive could be failing, and you would never know until you attempt to access it. Two such failures simultaneously mean you could lose the data on both failed drives. If you have a single failure, all other drives are immediately called into service, and must perform flawlessly. Regular parity checks confirm that a rebuild should go smoothly should one drive fail.

Ahhh thanks for the clarification.  I do spin up the drives but I don't know that I would recognize a problem.

 

Do you have any thoughts on a script to check parity on a monthly basis?

 

Also is there a tool I could use to schedule and  have proactive messages emailed to me in order to identify errors and other potential problems?

 

Thanks,

 

Neil

 

UnMenu has what you need.