I believe my MB has a faulty SATA Port ?

[lime111]

Hi Guys, it's been a long time since i've had a reason to post. 

 

In the past my 6th sata port on my MB had a brand new WD red die very soon after I bought it, it seemed odd at the time.  I decided not to use that port and bought a SATA expansion card as I needed extra expansion space for a cache.  

 

Just recently I have needed to expand my array, and my twin sata expansion card was full.  I upgraded my existing, working, parity with a new 8tb, and put the previous parity into sata port 6.  

 

It initially worked, but then had multiple crc errors pretty soon afterwards.   I've checked the server today and the disk in sata port 6 is disabled and being emulated. I've replaced the cable, no change.  I'm now not even sure the disk is truly dead.

 

Given multiple drives have failed on this one port, i can think it could only be the sata port, or the power into it, or am i missing something?

 

It's a bit annoying that I can lose another working 4TB drive, and i have absolutely no idea as to why/what is happening.  The only other variance with the array is disk 6 is formatted xfs, and every other disk is reiserfs.

 

I normally don't go first to hardware faults, but this time i think it might actually be the case.  I'm reluctant to buy another 4tb and would prefer to maybe buy a 4 port expansion card and never use sata port 6 again

 

Any input or advice appreciated. 

 

 

tower-diagnostics-20180828-1940.zip

[testdasi]

CRC error is almost always either cable or port fault and rarely ever disk fault.

You really have no choice but to get a bigger expansion card to test it out unfortunately.

[lime111]

Thanks @testdasi

 

I have bought a new SATA card, startechcom PEXSAT34RH , and managed to rebuild disk 6 and the array has started.  So I do believe the problem was the sata port on the MB.  However I now cannot start docker, or even upgrade any plugins.   I've deleted the docker image in case it got corrupted, but that didn't make a difference and the array just takes forever to start services, and I think eventually it just times out.  Thoughts...?  Log attached, thanks...  

tower-diagnostics-20180903-0755.zip

[John_M]

Disk 2 (3 TB WD Green) is failing. You need to replace it.

[John_M]

Also communication with your cache disk is being lost. So you still have a port or cable problem.

[lime111]

Thanks guys, ill sort out a new drive and new cable.  It hasn't been a cheap month for my tower.... 

[lime111]

Replaced failing drive and put in new SATA cable for cachce, array has returned to its old fine form, thanks again for helping out.