Hey ho!

 

/update LOG FILE INCLUDED IN POST FURTHER DOWN.

 

 

I'm having the weirdest issue where it is always Disk 4 in my Array that goes tits up.

 

• First it was a WD 8TB Ironwolf. 
• Then it was a 8TB Ironwolf Pro
• Then it was a 8 TB Ironwolf. (Not the same one as the first one.)
• Then it was a 10 TB Toshiba

 

Thing is: it is ALWAYS the disk that is connected in the Array as Disk 4. Physically they've shared almost nothing. OK, the power supply have been identical for all of them. But controller, cables and physical placement have all varied. 

 

The drives work fine in other arrays, and even in other (computer) hardware. I am totally perplexed

 

I'm suspecting Gremlins. Or really really low odds. But in reality I have no idea on what to do, or how to tackle the issue. I don't even know where to begin!

 

Any idea (or even pointing and laughing) is welcome!

 

 

TIA

 

/Akely  

 

Edited August 21, 20232 yr by Akely
Speling errör, log file update

Please post the diagnostics.

40 minutes ago, JorgeB said:

Please post the diagnostics.

That's probably something I should have thought of myself. Thanks for the pointer and the link. 

 

But will this help since "You need to do so BEFORE YOU REBOOT so that the logs show went wrong BEFORE the reboot (because once you reboot, it's lost)!"?

 

Yeah, I did that. /facepalm But I will not if it happens again. (So hopefully not.)

10 minutes ago, Akely said:

But will this help since "You need to do so BEFORE YOU REBOOT so that the logs show went wrong BEFORE the reboot (because once you reboot, it's lost)!"?

You could try enabling the syslog server with the mirror to flash option set to get a syslog that survives a reboot.

The persistent log server seems like something I will set up. Probably to a remote machine as I suspect that can be more useful if/when the Unraid server borks up. 

 

Thanks! 

Well, it happened again.

 

There definitely seems to be something really off as the thing that I *think* triggered the Disabling of Disk 4 was me hitting the "Compute All" button on the Shares page. The system seemed to get stuck, so I clicked Main then Shares again and did a share-by-share Compute. Then I got the error message.

Log file included, I'll also go through it. But I expect I'll understand little...

nas03-diagnostics-20230821-1905.zip

There are issues with almost all devices connected to the Asmedia controller, that controller uses SATA port multipliers and because of that is not recommended, see here:

 

13 minutes ago, JorgeB said:

There are issues with almost all devices connected to the Asmedia controller, that controller uses SATA port multipliers and because of that is not recommended, see here: <link deleted by Akely>

Thanks for the work JorgeB! 

 

Ouch! That could be a problem for me. I need to read up a bit and will find a way to circumvent the issue. As the system have been stable for ... a long time I suspect (but am not certain) that the issue could have started popping up when I added more drives to that card.

I'll look around for other cards, as per the recommendations in that thread.

I will make sure to report back. In the meanwhile other suggestions I can check up is also welcome. 

 

 

So I got a new SATA controller as per the thread JorgeB linked. Specifically it was a https://www.startech.com/en-se/cards-adapters/8p6g-pcie-sata-card.

I installed it and ripped the old one out. The Array is now rebuilding and time will tell if this is the issue or not. so far I've only noticed a slight change in spin up/down behavior. But if that is correlation or not I do not know.

 

//EDIT:

After a few days I think it's safe to say that the new card, with no SATA splitters on it have solved the problem. I marked the post from JorgeB as Solution.

Edited September 2, 20232 yr by Akely
Final(?) report and Solution marking.