Kernel panic - New Motherboard.

[sully]

So... I pulled the trigger with UnRAID based on the impending release of 6.2 (for the dual parity).  Setup a system under 6.1 (stable) using an old motherboard.  I've not used Unix/Linux for a good few years (15+) and this was going to let me ease back into what I had enjoyed using in College.  Everything was going well, I was getting comfortable with it, was about to see about extensions and dockers, but then something went awry.  Board (or UnRAID) would just stop at night during or after the Mover operation.  This was about two-three weeks into having the rig setup.

 

Troubleshooting of the issues inferred to me that it was the board (it failed to post with the disks unplugged and also with memory removed).  TBH, I started with hardware and when I couldn't get it to work, I decided to replace the board and get something with SATA3 native and PCIe v2 and v3 for expansion.  So if there is something wrong with the underlying UnRAID software setup then I don't know.

 

Ended up with a Gigabyte F2A88X-D3HP with an A10-7860K.  Another confession, I didn't check to see if this was on the blacklist (if there is one) for UnRAID, I figured the software is hardware agnostic and this wasn't that new a MB (I've googled the MB and searched the forums, at least one user has been successful in getting UnRAID setup with this board [of course they claimed they had no problems :-) ]).

 

So, a few items, that I'm wondering might solve my problem...

 

1.  I've enabled IOMMU, I think this was the cause of the Kernel panic when I first tried to boot.

2.  I've not attached the Disks yet.  I'm thinking this may be the cause of the subsequent problem(s), but again, I'm out of my depth RE: troubleshooting Linux.

3.  If I've understood the error message correctly this is the system's current gripe (or at least this is the last gasp from the Kernel before it halts):

 

    'not syncing: VFS: Unable to mount root fs on unknown-block(0,0)'

 

This seemed weird to me, as I thought UnRAID lived it's life completely off of the USB, so having the drives attached or not, shouldn't make a difference.  Or did I miss something, since this was a configured system with 12 data disks, two cache drives and one parity (with room for the 2nd).

 

I've gotten a different error messages on a prior boot, just after I changed the IOMMU setting in the BIOS, one mentioning something about looking at "linux documentation/init.txt for guidance".

 

All seem to point to the USB being angry that the system has changed and that it might need some configuration/settings adjustment that is, quite frankly, beyond me at present.

 

Also, I have no idea how to capture the boot sequence to disk or screen capture to share here, so if that's the next step, please be kind and provide some advice on how to complete that procedure.

 

Much thanks!

 

Sully.

[sully]

Looking a little harder at the video I made of one of the boot sequences, I think there may also be an error related to the /init.

 

"Failed to execute /init (error -2)"

"Kernel panic - not syncing: no working init found.  Try passing init= option to kernel.  See Linux Documentation/init.txt for more information" 

 

Anyway thanks for reading...

 

Sully

[sully]

Have tried several more things to resolve this issue, so I can boot into UnRAID.

 

1.  I've changed the UEFI BIOS Setting for the Flash Drive from UEFI to Legacy (the old board wasn't UEFI).

2. I've added the remaining hardware from the old machine (4 SYBA x4 SATA 3 Expansion Cards) and connected the Backplanes w/the 3TB Drive installed.

 

After completing #1, I still got the Kernel Panic message with the Init issues.

 

After completing #2, I didn't see any change in behavior.

 

Here's a pic of the end result (where the system hangs)...

 

Looking for some advice... At this point, I'd even accept someone telling me I'm a Muppet and should be able to figure this out on my own.

 

Thanks!

 

Sully.

[gatorguy]

What did you end up finding out?  I'm in about the exact same boat right now.

 

Background: I've had this unRAID box running 3 years or so with no issue.  I just recently did the latest update to 6.2.4, I think it was from 6.2.2, I think.  No hardware has changed in many, many months.

[John_M]

Have you tried checking your USB flash device for corruption? It might be worth re-making it from scratch? Save your config folder somewhere safe and download the installation zip from https://lime-technology.com/download/ then follow the instructions and see if it boots. If it does, shutdown and copy your config folder back from its safe place.

 

[sully]

I did just about everything I could think of, troubleshooting wise, and finally started playing with the memory timings and MemTest.  Scaled it back to the minimum the board was advertised to do.    Ran the tests for a day or so, then it booted up and everything was perfect.  For a short while...

 

The power went out about two weeks later and somehow, even though I invoked 'shutdown' form the command line (I had a UPS after all, just not setup in the GUI), it zapped the array config on the flash drive.  Messed up all the settings and the disks ended up back in a different order.  Not a huge deal, but I had planned on managing the disks/shares in a different way.  My fault, I suppose, as I hadn't remembered to update my data on which disk was which when I swapped MB's.  All my diagrams/labels were from the prior hardware config.

 

Before I gave up on the original config, I did run tests on the flash to see if it had any file system issues, but all came up clear.

 

Now it appears one of my brand new iStarUSA backplanes has precipitated a read/write error which has flagged a disk as faulty.  Since I don't have any spare at the moment and I'm about done with computer BS in general, I've shutdown the system till I feel like dealing with it, or I see a good deal on 3TB HD's...  i need one to replace the supposed faulty disk and a second for the dual parity.  I'll also replace the backplane with the spare I do have and RMA the one that's giving me problems. 

 

But since I need to decide on what disks and at what price (who knows how long), order the disks (up to a week for shipping), preclear them (3-4 days), then rebuild the array (who knows how long, since I've not done it before), I'm just not in the mood anymore.

 

It's my luck that I'll start having other HD errors on other machines, then it'll be a perfect storm of failure for my UnRAID plan...

 

Perhaps I was too ambitious... Should have stuck with building it in one of the HP Microservers with 6 disks instead of going for 16, but that just wasn't the density I was after with dual parity.

[gatorguy]

I followed the instructions in the following thread to resolve my issue!

 

https://lime-technology.com/forum/index.php?topic=52864.0