Hardware issues

[crazystar]

Hello all,

 

I am running quite a few disks in an R710 hooked up to two SA120s(daisy chained using SF-8088 cables). I am getting quite a few CRC errors...after reading through the forum, it looks like a data cable issue(?). Now these cables cost $25 each so before I drop $50 trying to debug, is there anything else I can do to narrow down the issue?

 

Any help is appreciated.

 

Thanks

rangoonmediasrv-diagnostics-20180715-2100.zip

[Frank1940]

Suggestion.  Why don't you go through all of the SMART reports in your diagnostics file and make a list of the disks that the disks have the CRC errors. Then using that list determine exactly what hardware is involved with those disks. 

 

I suspect that you will find that there will be some piece(s) of  hardware that is/are common to all of the disks with CRC errors.    

[crazystar]

13 hours ago, Frank1940 said:

Suggestion.  Why don't you go through all of the SMART reports in your diagnostics file and make a list of the disks that the disks have the CRC errors. Then using that list determine exactly what hardware is involved with those disks. 

 

I suspect that you will find that there will be some piece(s) of  hardware that is/are common to all of the disks with CRC errors.    

 

Looks like the devices are all in SA120s...I guess I buy the SF-8088 cables now?

[Frank1940]

In both SA120 racks or just one?  Over what period of time did these errors to occur?   Remember a CRC is not a hard failure.  It simply says that the data that came out of the end of that cable was not what was sent.  When that occurs, the data will simply be resent which if the errors are widely spaced out is virtually unnoticeable.  Plus, the error count is from the day when you first put power to them.  'Fixing' the cause will not change the count back to zero, it simply would not increase beyond the current count. 

 

I find it a bit difficult to believe that both cables are bad.  Be sure that you can't isolate the problem a bit more.  Also consider that the cables could have electrical noise coupled into them since they are not inside of a metal case which would help to shield them.  

[crazystar]

15 minutes ago, Frank1940 said:

In both SA120 racks or just one?  Over what period of time did these errors to occur?   Remember a CRC is not a hard failure.  It simply says that the data that came out of the end of that cable was not what was sent.  When that occurs, the data will simply be resent which if the errors are widely spaced out is virtually unnoticeable.  Plus, the error count is from the day when you first put power to them.  'Fixing' the cause will not change the count back to zero, it simply would not increase beyond the current count. 

 

I find it a bit difficult to believe that both cables are bad.  Be sure that you can't isolate the problem a bit more.  Also consider that the cables could have electrical noise coupled into them since they are not inside of a metal case which would help to shield them.  

 

So I looked the CRC error count and it looks like the counter is going up for drives in one of the SA120s. I will replace that cable and report back. 

 

Btw, is there anyway to clear these counts?

 

[Frank1940]

Just now, crazystar said:

Btw, is there anyway to clear these counts?

 

No, they are stored internal to the drive.  IF you move the drive to a new computer all of the SMART data (including the CRC error count) will move with it.