What does SMART ATA error mean exactly?

danizmax wrote:
...So what do exactly the following mean: Current Pending Sector Offline Uncorrectable ATA Error

There is a general explanation here: http://en.wikipedia.org/wiki/S.M.A.R.T.

Current Pending Sectors: Those are sectors that could not be read. The NAS would have to reconstruct the data from the redundant part of the raid array.
Offline Uncorrectable: The total count of uncorrectable errors when reading/writing a sector. In your case these are due to the pending sectors
ATA error: The SATA controller on the ReadyNAS was unable to communicate with the hard drive.

Check on the seagate site to see if your drive firmware is down-rev. https://apps1.seagate.com/downloads/req ... kie=en_US_
You will need the drive serial number and part number to do the search. If it indicates that your drive firmware needs updating, you can do that in a Windows PC.

Thanks for the quick answer.

StephenB wrote:
ATA error: The SATA controller on the ReadyNAS was unable to communicate with the hard drive.

The log is from the HD right? So isn't the other way around then, that HD had problem communicating with the controller on the NAS?
I have no idea, just asking to confirm ;P

Thanks!

I believe that the ATA error count was added by Netgear, and is instrumented in the NAS (unlike the real SMART parameters which are instrumented/stored in the drive itself).

danizmax wrote:
why does my HD have so many flying head hours? Head Flying Hours: 99291053949400

The time is stored as a 48-bit number, but only the lower 32 bits hold the hours; the upper 16 bits hold a running count of milliseconds and should be ignored.

Your number, 99291053949400 decimal, is 5A4E 0000 01D8 hex. The "0000 01D8" is the actual number of hours -- 472 decimal, which sounds about right for an 8-month-old drive that spins a couple hours a day. My drives spin continuously, and one of my 8-month-old drives reports 180504590553978 flying hours. 180504590553978 decimal is A42B 0000 137A hex; 137A hex is 4986 decimal, which works out to 207 24-hour days.

Thanks for the explanation, I just wonder why don't they just recalculate for the user.
You could always recalculate back to decimal if you really need it, but I doubt anyone would xD

danizmax wrote:
Thanks for the explanation, I just wonder why don't they just recalculate for the user. You could always recalculate back to decimal if you really need it, but I doubt anyone would xD

Some of these fields have vendor-specific formats.

OK, so I've patched the hard disk and reformatted, but still got 6 bad sectors and ATA errors. I just need to be sure that the controller on my NAS is fine before I proceed with the HD waranty.

So what happens if I switch the hard disks, so that HD1 is in slot 2 and HD2 is in slot 1? Can I test the controlled that way or is there any other way to test this?

The v2 is supposed to handle disks out of order ok (note I don't have one). Make sure you swap them with the NAS powered down, or you will wipe your data.

You can also double-check the SMART stats on a PC (you might not see ATA errors, but you will see the bad sector report). If you have scratch disk(s) you can also use them to test the controller, keeping your real disks safely intact. As noted above, hot-inserts will wipe the drives, so be careful on that.

The v2 boot menu includes an off-line full disk test (described here: http://www.downloads.netgear.com/files/ ... Mar-12.pdf). If you use the scratch disk idea, that should also exercise the controller. I'd avoid running the diag on the disk which has known bad sectors right now.

BTW, if you are in the US, Seagate has an "advanced replacement" option on the RMA - where they send you the replacement first, with a return shipping label for the old disk. The price is quite reasonable (about the same as I would have paid for shipping anyway). You get the replacement much quicker. That's one way to get a scratch disk (since you have a bit of time before you need to send the defective disk back).

The RMA might not be on the HCL, so you might also consider buying a new drive, and using the RMA somewhere else - perhaps as an emergency spare.

So I swapped the hard disks, and I see same problem in bay one instead of bay two now, and even NAS recognized it's the same disk, which can only mean the controller is fine.
I'm not in US which unfortunately means I'll have to do more convincing than you guys and it usually costs more. I hope these logs and 6 bad sectors will be enough.

Anyways thanks for all the info and help!

Regards, Daniel

If you are taking the disk out of service anyway, you can also run the long read and write tests in Seagate's Seatools diagnostic.

If there are more unwritable sectors, the write test should find them reallocate them. It is a destructive test though.