Seagate st3000dm001-9yn1 smart attribute [b8]

Question

Seagate st3000dm001-9yn1 smart attribute [b8]

The status of this smart attribute is showing: failed b8 Unknown Attribute 99 30 30 FAILED

Where can I find more info about the 'unknown' [b8] attribute?

The disk is connected via highpoint raid controller so SpeedFAN won't read it's s.m.a.r.t to lookup in online hddstatus database.

Answer 1

Wikipedia's list of known SMART attributes lists 0xB8 as End-to-End error / IOEDC described as:

This attribute is a part of Hewlett-Packard's SMART IV technology, as well as part of other vendors' IO Error Detection and Correction schemas, and it contains a count of parity errors which occur in the data path to the media via the drive's cache RAM.

(I was unable to immediately find an official reference of SMART attributes for your particular drive, so will assume that the above is correct. You may want to spend a bit of time searching for a specific reference which will tell you exactly what the meaning (and interpretation of the raw value) of SMART attribute 0xB8 is on your particular drive, to verify the below. Seagate should be making such a reference available.)

Parity errors can be caused by hardware being operated too close to the limit of its operational envelope (overclocking), but a far more plausible explanation would appear to be poor connectivity somewhere.

The first thing I would do is check the cabling. Replace the cabling from the controller all the way to that drive (including any breakout cables) and see if the problem goes away.

Note that parity errors are non-critical. This is a parity error which is being detected, and thus corrected for (most likely by throwing out the damaged data and requesting a re-read from the original source). SMART may read it as FAILED because the drive has seen too many such errors, but at most, detected parity errors degrade performance. As long as this is the only error that the disk is reporting, I wouldn't worry too much.

---

It's also worth noting that the Seagate DM series drives are a poor fit in general for RAID because they don't support SCTERC. As long as they are working properly this won't be a problem, but if you hit an unreadable sector, the drive is going to keep trying to read it for a very long time even though (depending on your RAID level) redundancy may exist elsewhere to allow it to be reconstructed much quicker. That applies to any RAID level other than RAID 0 (which is striping without mirroring or parity).

Answer 2

I assume this is for an HP computer as B8 is a special firmware revision change for Seagate drives. Basically it monitors the data from the point it is loaded into the HDD cache and travels to the Host controller. If the checksum does not match then an error will occur which is recorded in this attribute. This means two things..

1. The cache on you HDD's PCB is failing (I don't believe this is the case)

2. There is an instability in your motherboards storage controller (this is more likely, are you overclocking, etc? Try updating you drivers.)