Need advice for dealing with hanging problem strategy

[Printingdude]

My UnRAID is an old v. 4.5.6, with 6 two-Tbyte drives, for 10 TB + Parity.

Lately it hangs and goes offline.  I restart, it comes back online, and I wait several hours for parity to validate (it does), and my drives and data show up intact and without any errors or faults. But when start to save some data to it, it hangs and goes offline within a few minutes of copying data.

I've repeated this process several times to convince me that it isn't a problem that is going to go away.  I'm trying to work out my best strategy to overcome this problem.

 

Back in 2010 when I bought my UnRAID license, I bought a Server-Pro registration 2-pack, of which I've only used one registration.

Reading a bit about how to troubleshoot v.4 problems, one of the first pieces of advice is to upgrade to a current version.

I see it is recommended to do a "clean install" that requires reassigning drives, settings, users, and passwords; but should retain my data. 
I run my UnRAID headless and keyboardless, but can attach a monitor and keyboard if needed for troubleshooting.

I see I can set up a system log monitoring telnet session, as well, to help troubleshooting, since the thing is hanging and going offline.  No PING response, no web login, it's out-to-lunch.

My questions:  Is it worth spending time capturing logs and seeing what it's reporting, if anything, when it fails?  –or–

Should I just go for upgrade to current version?  –and if so– should I set aside my USB stick drive and start a new one with my other registration code, just so that I don't actually damage or change anything on my current 4.5.6 install, so that I can at the very least get back to exactly where I am right now if anything goes wrong with the clean install?  (This strikes me as a wise option.)

 

Any tips and advice welcome.

 

[JonathanM]

First thing that comes to mind is hardware faults, which won't get fixed by upgrading software.

 

Can you get smart reports on all your drives?

Have you run memtest for an extended multipass run? (12 hours or so)

Have you cleaned out the dust bunnies recently and checked for fan operation?

 

Do you have backups of all the data on the array that you wouldn't want to lose? If not, copying data to backups would be first priority.

[Printingdude]

I cleaned out the dustbunnies (and there was a considerable bit to clean out).

I haven't done any other testing.  I'll give those a shot and see what turns up.

I likely do *not* have backups of the data.  I'm worried that with the system hanging, it will do that if I attempted to copy data to a networked system, although I *have not tried* yet.

I'd almost like to have another UnRAID system to back up the data to.  Now that hard drives are even larger and cheaper, I may put together a second UnRAID just for that purpose.

[JonathanM]

3 hours ago, Printingdude said:

I'd almost like to have another UnRAID system to back up the data to.  Now that hard drives are even larger and cheaper, I may put together a second UnRAID just for that purpose.

That would be hands down your best option. Get a new server up, then start copying data.

 

Reading data from the server is less stressful than writing to it, which is what you stated caused the hang.

[Printingdude]

Got a new 16TB 3-drive unRAID up and running with the latest version.
Tried to copy one directory of content and unRAID #1 dropped offline again.
I'm suspecting this might be a power problem and may swap out the PSU before trying again.

[Printingdude]

Update:  My hunch about the power supply appears to be correct.  Replacement with a new Corsair CX650M appears to have solved the problem.  Now backing up content to the new 16TB unRAID.