Let me start out by saying I'm a total "newbie" as far as unRAID is concerned.  I am attempting to set up an unRAID server.  Initially I thought things worked pretty well.  I could write to the array, read from the array and set up some virtual machines.

 

Now, though, just several days (and multiple attempts at resetting the array) later, every single time I try to write to the array, things come to a crashing halt.  The virtual machine stops dead in its tracks.  When I try to look at a directory listing of what's in the array, I can see some directories but the report shows no files.  The only way I can get the array to work again is to shut the entire system down and reboot.  And then it works until the next write attempt.

 

I have looked at the diagnostic report (which is attached) and I'm wondering if the following might be significant:

 

smartctl 6.5 2016-05-07 r4318 [x86_64-linux-4.9.7-unRAID] (local build)

Copyright © 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

 

ERROR: smartctl takes ONE device name as the final command-line argument.

You have provided 2 device names:

 

/dev/sdj

 

Use smartctl -h to get a usage summary

 

smartctl -h

 

I have no clue at all as to how I have provided 2 device names.  And, completely unrelated, I have also come across a statement somewhere that I have two registration keys!!  I've only paid for one, so how I have two of them is a complete mystery.

 

In any event, I don't know what to do about either error.  I'd certainly appreciate someone wiser than I in matters of unRAID pointing me in the right direction!

 

Thank you in advance!

 

John Snyder

z9pe-unraid-diagnostics-20170207-1224.zip

Looks like you still have your Trial Key on your flash in addition to the Basic Key, so that is probably what that is about.

 

All your disks are having read errors. Possibly a controller problem. Tell us about your hardware.

 

The SMART problem is a bug in unRAID 6.3.0. See here: https://lime-technology.com/forum/index.php?topic=56204.msg536774;topicseen#msg536774

 

Feb  7 12:22:16 Z9PE-unRAID kernel: ata4: hard resetting link

Feb  7 12:22:17 Z9PE-unRAID kernel: ata3: SATA link up 3.0 Gbps (SStatus 123 SControl 300)

Feb  7 12:22:17 Z9PE-unRAID kernel: ata4: SATA link up 1.5 Gbps (SStatus 113 SControl 310)

 

Looks like SATA cable problems. Power down and check cables. Normally I'd ask you then to power up and grab new diagnostics and post so we can see if you have any SMART problems but the bug means that SMART reports gathered by Tools -> Diagnostics are empty. So you'll have to check each disk's SMART report via the Main page (click the disk name) or Dashboard page (click the thumb or triangle) of the GUI.

 

I have opened up the case and checked the SATA connections - and they appear fine. 

 

I just did a SMART test on all 4 hard disks with the computer not in an error state.  They all "completed without errors".

 

I have also attached a diagnostics report with the computer not in its error state.

 

My hardware is as follows:  ASUS Z9PE motherboard, dual Xeon 2687W processors, 32 gigs of RAM, 4 HGST 6TB NAS hard drives, 2 1GB Samsung SSDs.

 

The computer worked properly until right around the time I upgraded to unRAID 6.3.  I don't remember the exact time frame, but the error condition could well have started with the upgrade.

 

Question:  how do I get rid of the unwanted trial key?

Question:  how do I correct the issue regarding the hard disks having 2 device names?

Question:  what would be the best way to downgrade from unRAID 6.3 to the previous version?

 

Thanks.

z9pe-unraid-diagnostics-20170207-1430.zip

I received this email from the unRAID server immediately after it went into its error condition:

 

Event: unRAID Parity disk error

Subject: Alert [Z9PE-UNRAID] - Parity disk in error state (disk missing)

Description: No device identification ()

Importance: alert

 

I received 4 identical emails, one for each of the 4 hard disks.

You didn't mention your power supply, but since you don't have many disks yet that is unlikely to be the problem unless it is defective. Is it Single +12V Rail?

 

The key files are in the config folder of your flash.

 

Since you don't have one of the controllers some have had issues with, it does seem like it may be a connection problem. Check both ends of power and sata cables to the disks. If you have bundled your SATA cables you might consider letting them loose. Bundled cables can put stress on the connection and cause them to work loose or get pulled so they aren't square on the connector.

Question:  how do I correct the issue regarding the hard disks having 2 device names?

 

They don't have two device names. It's a bug. I presume it will be fixed in 6.3.1.

 

Question:  what would be the best way to downgrade from unRAID 6.3 to the previous version?

 

Assuming you upgraded from the previous version, replace the three bz* files in the root of your boot flash with the three files in the previous folder on your boot device.

The power supply is a CORSAIR AX Series AX860 860W 80 PLUS PLATINUM Haswell Ready Full Modular ATX12V & EPS12V SLI and Crossfire Ready Power Supply.

 

I have checked the SATA cables and the power cables.  They are not bundled together.

 

Thanks for the info on the key files.  I'll look and try to delete one of them.

I did replace the three bz* files with those from the previous installation.  unRAID booted up, but I lost my virtual machines and my shares.  So, without a lot of further investigation, I reversed the above and replaced the bz* files from the 6.2 installation with those from the newer 6.3 installation.

 

Right after I reboot, I access the file system over the network from my Windows 10 computer, I see my files.  I can read them and I can write to the unRAID disk. 

 

My problem occurs when I start a virtual machine and try to do anything that involves a disk access - like install a program or an update.  The virtual machine just stops dead in its tracks.  When this occurs, I cannot access my backed up data on the unRAID server; it doesn't even show.  I can see the server, I can see my shares, but any data inside the shares doesn't show.  I have to reboot the unRAID server in order to once again see my data.

 

Alternatively, instead of rebooting, I have taken the array off line and attempted to restart it.  At that point, I get an error message saying "Stopped. Invalid configuration. Too many wrong and/or missing disks!"  When I booted up with the bz* files from the 6.2 version of unRAID, I could stop and start the array at will; this error did not occur.

 

Does this information shed any light on the problem???

Does this information shed any light on the problem???

 

Not really, but your latest diagnostics might. It looks as though you have a problem with whichever disk is hosting your VMs and possibly others. We need to see SMART reports but the bug in 6.3.0 prevents them from being included in the diagnostics zip. Diagnostic would also tell us if there are still communication problems between controller and disks. Looking at the SATA cables isn't enough. You should at least reseat them and preferably replace them.

 

I recommend not trying to use VMs until the disk issues are resolved.

 

I'll get some new SATA cables for all 6 drives.  I'll let you know what happens.

I don't presently have anything on the array that I can't afford to lose.  So I decided to erase the entire flash drive with the exception of my key and copy the latest version of unRAID to the flash drive.

 

I did that - I was able to mount all the disks properly and I didn't loose any data.  Additionally, all the array problems disappeared, and everything is working properly just as it did when I first began.

 

So ... I don't think it was a hardware problem.  I suspect that I inadvertently "tweaked" some setting and screwed things up!  And ... I'll probably do it again as I learn about unRAID!!