New 2.00.04 Smart info error

[crazybucky]

I noticed under the new firmware 2.00.04 that new SMART info had been added to the Disks Menu.

My problem is that the info for the disk seems incorrect?!

This Drive, I get a Warning message because of the Reallocated Sector Count but should the Raw Read Error Rate, Seek Error Rate, Hardware ECC Recovered be so high?

Tray Number 1
Model ST31000340NS
Power On Hours 774 Hours
Temperature Celsius 40
Reallocated Sector Count 3
Current Pending Sector 0
Raw Read Error Rate 114546622
Seek Error Rate 9924472
Hardware ECC Recovered 114546622


This Drive is fine...
Tray Number 5
Model ST31000340NS
Power On Hours 746 Hours
Temperature Celsius 40
Reallocated Sector Count 0
Current Pending Sector 0
Raw Read Error Rate 119428947
Seek Error Rate 9857275
Hardware ECC Recovered 119428947

Looking at it, the Raw Read Error Rate and Hardware ECC Recovered have the same values. I know this isn't a "supported" drive. Anyone else having trouble?

[cyrilix]

That is strange. I have 5 drives in my N5200, and the first 4 were put in at the same time, the last one was put in later. The first 4 all have Power On hours within 1 hour of each other, no raw read and seek error, so your number does seem quite messed up, although it's possible that there have been many errors on your drive. Do you have md5 hash any of your files? If so, maybe run a check to see that they still pass the hash check.

[N2100Owner]

I wonder if the drive will report the same SMART information if you connect the drive to a Windows box (don't format the drive of course)? I personally use Lavalys' Everest.

[crazybucky]

I can explain the power on hours. I got all 5 drives from newegg at the same time. But I did notice that one drive did have like 4 hours already on it even before I started. So I did hooked it to my computer and ran the Full seagate tool test (took quite a while). but all the smart and test reasults came back fine. I too even run everst and the results where fine.

I doubt my drives are bad but the reason I asked was to see if anyone else had weird smart errors on their drives or maybe it's just mine seeing as the drives aren't supported, possible bug, or maybe since the drives I have are server class?!

I'll post my other drives info, because I doubt I have 5 server grade hard drives going bad on me, least I hope! ;D


Thanks for your feed back!

[crazybucky]

Tray Number 1 Model ST31000340NS Power On Hours 793 Hours Temperature Celsius 39 Reallocated Sector Count 3 Current Pending Sector 0 Raw Read Error Rate 143950014 Seek Error Rate 10408985 Hardware ECC Recovered 143950014

Tray Number 2 Model ST31000340NS Power On Hours 764 Hours Temperature Celsius 41 Reallocated Sector Count 4 Current Pending Sector 0 Raw Read Error Rate 66716268 Seek Error Rate 11148328 Hardware ECC Recovered 66716268

Tray Number 3 Model ST31000340NS Power On Hours 764 Hours Temperature Celsius 43 Reallocated Sector Count 0 Current Pending Sector 0 Raw Read Error Rate 73637320 Seek Error Rate 10221222 Hardware ECC Recovered 73637320

Tray Number 4 Model ST31000340NS Power On Hours 764 Hours Temperature Celsius 43 Reallocated Sector Count 0 Current Pending Sector 0 Raw Read Error Rate 89406435 Seek Error Rate 10265636 Hardware ECC Recovered 89406435

Tray Number 5 Model ST31000340NS Power On Hours 765 Hours Temperature Celsius 39 Reallocated Sector Count 0 Current Pending Sector 0 Raw Read Error Rate 130604142 Seek Error Rate 10344883 Hardware ECC Recovered 130604142

[cyrilix]

From what I see, your Hardware ECC Recovered is the same as your Raw Read Error Rate, which sounds to me like every read error has been somehow mitigated. It may simply be that the drives themselves give off different SMART data or that for this specific drive, those bits of data are not really indicative of anything important. That's the best hypothesis I could come up with.

For the record, mine has 0 errors, and under Hardware ECC Recovered, it says N/A so maybe for my drive, those fields are only used for actual errors, whereas for your drive, those fields are used for some other common hard drive access operation.

[fajo]

This actually is an issue with smartctl and Seagate disks - Seagate want their seatools to be used obviously.

The values for

• temperature
• power on hours
• power cycle count (not shown in WebUI)
• start/stop count (not shown in WebUI)

seem to be interpreted correctly by smartctl - all the rest should be ignored since smartcrl would have to interpret the normalized values to correctly show it.

/Falk

[sgwarez]

Hi,
My first disk also have warning ! But no problems with the other 4 disk! Last night I resolve remove the disk, rum the sea tools, and no problems, reported! More I decide make a full erase, and after re-install on my NAS, and the system rebuild the raid but the problem persist! Any idea how resolve the problem ?

Tray Number 1
Model ST3320620AS
Power On Hours 1521 Hours
Temperature Celsius 35
Reallocated Sector Count 1
Current Pending Sector 0
Raw Read Error Rate 238875428
Seek Error Rate 85704939
Hardware ECC Recovered 176122351

[Georg Eschenbach]

Hi,

yes, the funny thing is, that on the seagate drives
( I am running ST 3100 340NS)
is the RAW Read Error Rate = Hardware ECC Recovered.

On a second N5200 with WD 1000 FYPS there are
RAW Read Error Rate = O and Hardware EDD recovered N/A.

It really seems that the SMART info is not a really good information.

Greetings
George

[fajo]

Apr 6, 2008 19:08:44 GMT 7 Georg Eschenbach said:

It really seems that the SMART info is not a really good information.

... at least not for Seagate disks since the values for
"Raw Read Error Rate", "Seek Error Rate" and "Hardware ECC Recovered" are not interpreted correctly by the smart tools (they are not linear as smartctl seems to believe). This is not a bug in smartctl but rather caused by Seagate not publishing information regarding this attributes.

Check if your disk model is in the smart database - if not you may not want to trust the values displayed.

/Falk

[lordofschnitzel]

Ive got these crazy smart ecc seek and read errors too on my 8x 750GB seagates. Just popped in a brand new 1TB seagate also and it reports a few thousand errors - not possible as its new. Just wrong errors.

On teh flip side, can we please get this fixed in the next firmware release, so we actually KNOW when theres real errors?

[fajo]

Apr 20, 2008 17:56:33 GMT 7 lordofschnitzel said:

On teh flip side, can we please get this fixed in the next firmware release, so we actually KNOW when theres real errors?

This can't be fixed in the FW since it is smartctl which cannot interpret these values. Thecus can only stop showing them on the UI.

/Falk