Dear Community,

 

I am writing you to get every possible sort of support about a strange fact that happened me today.

 

unRAID version 6.2.3. Diagnostics output attached (thanks trurl).

 

Until yesterday my unread machine, built and initialized last week, was running quite good.

The setup was made of 4x 3TB seagate drives and 3x 5TB WD Red drives (formerly from a WD MyCloud Ex4) as data disks, 1x 4TB WD Red drive as parity drive and 3x 500GB drives merged into the cache pool.

Everything was running well, given that the data drives as well as the parity one were connected to a re-flashed Dell H310 raid card using a pair of SAS->4xSATA cables.

The cache drives were installed directly on the motherboard SATA headers.

 

Today I have switched on the system for further settings and optimizations and 2 drives were not detected.

On the syslog I have observed a strange message:

mtp2sas: log_info(***): originator(PL), code(0x11), sub_code(0x1000)

Happening immediately after the initialization of other drives.

 

At first I was thinking about a H310 failure, but what I have discovered after several attempts is that 2 out of 4 WD Red disks are now no more recognized neither by UnRaid (hanging condition reported above) or by the BIOS, avoiding the use of the expansion card.

 

If only one disk were faulty I would think that that was a hardware failure, but since 2 disks are now broken I can't figure out what happened.

 

I would like to specify that I can exclude a power glitch or surge since the two faulty drives were the 2nd and the 4th drive connected to the same power cable, and the 3rd drive is working perfectly.

 

Thanks in advance for your really appreciated help!

 

GF

 

alz101xl-diagnostics-20161111-1500.zip

See Need help? link in my sig.

Dear trurl,

 

Your post is rather cryptic.

If you wanted to tell me to provide the syslog and diagnostic output I will do that tomorrow, since actually I am far away from the system.

If, instead, you would tell me to run a SMART test it would be a little bit difficult to do: the devices are no more listed in de /dev directory!

Thanks for replying anyway

 

GF

Up, diagnostic data has been attached to the first post!

* Motherboard BIOS is from early 2008, check for a newer one

 

* Check also for a newer firmware for that SAS card

 

* SAS cards using the mpt3sas drivers are notoriously hard to troubleshoot (I almost wish users wouldn't use them!), as they don't provide anything helpful for troubleshooting.  The syslog says that the first 6 ports are in use, and identified and initialized them.  It saw nothing on the 7th port.  It saw something on the eighth port, but almost immediately aborted it.  No human intelligible clues as to why.

 

* You do have 3 unused ports on the motherboard, and since these are perhaps your best ports, why not move 2 or 3 of the drives to there.  (It's probably not a drive problem, but if it were, these ports give great troubleshooting info!)  If you reconnect the missing drives on the motherboard, then they should reappear in the array.  Tell us what you see.