[Seaside]

Most of you already knew there is a HDD monitoring function called SMART.
It monitors and analyzes HDD and will report you when the HDD is to be replaced. My question is how reliable this SMART thing is?

Here's a SMART reports by HD tune softwear. That's my old 120G HDD.



HD Tune: ST3120026A Health

ID Current Worst Threshold Data Status
(01) Raw Read Error Rate 56 52 6 109948709 Ok
(03) Spin Up Time 96 96 0 0 Ok
(04) Start/Stop Count 100 100 20 31 Ok
(05) Reallocated Sector Count 100 100 36 0 Ok
(07) Seek Error Rate 87 60 30 578491260 Ok
(09) Power On Hours Count 74 74 0 23202 Ok
(0A) Spin Retry Count 100 100 97 0 Ok
(0C) Power Cycle Count 100 100 20 362 Ok
(C2) Temperature 35 51 0 35 Ok
(C3) Hardware ECC Recovered 56 52 0 109948709 Ok
(C5) Current Pending Sector 100 100 0 0 Ok
(C6) Offline Uncorrectable 100 100 0 0 Ok
(C7) Ultra DMA CRC Error Count 200 199 0 1 Ok
(C8) Write Error Rate 100 253 0 0 Ok
(CA) TA Counter Increased 100 253 0 0 Ok

Power On Time : 23202
Health Status : Ok

Something bordering me with old HDD report is...

(01) Raw Read Error Rate 56 52 6 109948709 Ok
(03) Spin Up Time 96 96 0 0 Ok
(07) Seek Error Rate 87 60 30 578491260 Ok
(C3) Hardware ECC Recovered 56 52 0 109948709 Ok

I was like WTH, this dosen't make any sense. Spin up time of 0?
What's up with those error rates? Error rate of good drive is supposed to be 0. In fact, all my other HDDs error rate is 0.

It looks like everytime it read, an error occured. Error numbers are going up at this very moment. If the record is correct, the HDD is seriously messed up. But I've been using this HDD as boot drive for years without any apparent problem, so i thought this HDD is rock solid.

So... question is, is this SMART reliable?
Should i ignore the report or, is the HDD really screw**d up?

 

[003]

One time I got a SMART message when booting up my computer saying my hard drive was about to fail. Sure enough, it failed a few days later, I believe.

 

[LawnGnome]

Quote:

Originally Posted by Seaside /img/forum/go_quote.gif Most of you already knew there is a HDD monitoring function called SMART. It monitors and analyzes HDD and will report you when the HDD is to be replaced. My question is how reliable this SMART thing is? Here's a SMART reports by HD tune softwear. That's my old 120G HDD. HD Tune: ST3120026A Health ID Current Worst Threshold Data Status (01) Raw Read Error Rate 56 52 6 109948709 Ok (03) Spin Up Time 96 96 0 0 Ok (04) Start/Stop Count 100 100 20 31 Ok (05) Reallocated Sector Count 100 100 36 0 Ok (07) Seek Error Rate 87 60 30 578491260 Ok (09) Power On Hours Count 74 74 0 23202 Ok (0A) Spin Retry Count 100 100 97 0 Ok (0C) Power Cycle Count 100 100 20 362 Ok (C2) Temperature 35 51 0 35 Ok (C3) Hardware ECC Recovered 56 52 0 109948709 Ok (C5) Current Pending Sector 100 100 0 0 Ok (C6) Offline Uncorrectable 100 100 0 0 Ok (C7) Ultra DMA CRC Error Count 200 199 0 1 Ok (C8) Write Error Rate 100 253 0 0 Ok (CA) TA Counter Increased 100 253 0 0 Ok Power On Time : 23202 Health Status : Ok Something bordering me with old HDD report is... (01) Raw Read Error Rate 56 52 6 109948709 Ok (03) Spin Up Time 96 96 0 0 Ok (07) Seek Error Rate 87 60 30 578491260 Ok (C3) Hardware ECC Recovered 56 52 0 109948709 Ok I was like WTH, this dosen't make any sense. Spin up time of 0? What's up with those error rates? Error rate of good drive is supposed to be 0. In fact, all my other HDDs error rate is 0. It looks like everytime it read, an error occured. Error numbers are going up at this very moment. If the record is correct, the HDD is seriously messed up. But I've been using this HDD as boot drive for years without any apparent problem, so i thought this HDD is rock solid. So... question is, is this SMART reliable? Should i ignore the report or, is the HDD really screw**d up?

different companies use different values to denote good and bad. Look around and you can find some explanations on the different values.

I believe in SMART. As it can normally predict degenerative failures well. Of course it can't predict sudden failures from shocks or drops.

Remember though, SMART monitoring must be enabled in the bios for it to function properly. You can still read the values if monitoring is disabled, but I don't think they are updated.

 

[Bizzel]

Yep, I got a SMART warning on my old computer's hard drive and it failed just a month or two after. It can be a little bit alarmist at times but it's better to err on the side of caution.

 

[LawnGnome]

Quote:

Originally Posted by Bizzel /img/forum/go_quote.gif Yep, I got a SMART warning on my old computer's hard drive and it failed just a month or two after. It can be a little bit alarmist at times but it's better to err on the side of caution.

Very true, especially when you have a few years of irreplaceable data.

Part of the reason I want a RAID 1 setup.

 

[Fallingwater]

Quote:

Originally Posted by Seaside /img/forum/go_quote.gif HD Tune: ST3120026A Health ID Current Worst Threshold Data Status (01) Raw Read Error Rate 56 52 6 109948709 Ok (03) Spin Up Time 96 96 0 0 Ok (04) Start/Stop Count 100 100 20 31 Ok (05) Reallocated Sector Count 100 100 36 0 Ok (07) Seek Error Rate 87 60 30 578491260 Ok (09) Power On Hours Count 74 74 0 23202 Ok (0A) Spin Retry Count 100 100 97 0 Ok (0C) Power Cycle Count 100 100 20 362 Ok (C2) Temperature 35 51 0 35 Ok (C3) Hardware ECC Recovered 56 52 0 109948709 Ok (C5) Current Pending Sector 100 100 0 0 Ok (C6) Offline Uncorrectable 100 100 0 0 Ok (C7) Ultra DMA CRC Error Count 200 199 0 1 Ok (C8) Write Error Rate 100 253 0 0 Ok (CA) TA Counter Increased 100 253 0 0 Ok Power On Time : 23202 Health Status : Ok

The status for all of this stuff is OK. Smart seems to not have seen any malfunction. Did you run the scan yourself or did it start on its own accord?

Quote:

(01) Raw Read Error Rate 56 52 6 109948709 Ok (03) Spin Up Time 96 96 0 0 Ok (07) Seek Error Rate 87 60 30 578491260 Ok (C3) Hardware ECC Recovered 56 52 0 109948709 Ok I was like WTH, this dosen't make any sense. Spin up time of 0? What's up with those error rates? Error rate of good drive is supposed to be 0. In fact, all my other HDDs error rate is 0.

As stated, different brands have different ways of reporting data. Maybe the big number is just the number of read/write operations and the small numbers are the errors? I don't know, but I still see "Ok" after those numbers.

Quote:

But I've been using this HDD as boot drive for years without any apparent problem, so i thought this HDD is rock solid.

The fact that you've used it as your boot drive for years is valid information against it, not in favor of it. All hard drives slowly degrade and eventually fail. It's just a matter of time. The longer you've used it the more likely it is to drop dead.

 

[jeeger]

There's not much to believe. S.M.A.R.T itself is just a monitoring system for hard drives, with no analysis. Since the data is interpreted differently from manufacturer to manufacturer, you need some sort of database with the thresholds (on Linux see the manpage of smartctl for an explanation). These threshholds are something to "believe" in, and since they should be empirically determined, I assume them to be pretty accurate.

@Seaside: I think you are looking at the raw values.

 

[Seaside]

Of cause, SMART is enabled in bios.
Odd thing about that drive is, SMART still say it is ok even if it sees that much errors. I think there are two possibilities, 1. SMART see something else as errors, or 2. They actually are errors but they can be corrected internally without data loss, and there is no physical demage. So SMART says it's ok.
In fact, surface scan did not detect any demaged sectors.

I think I read help file of some monitoring program (i think that is hotcpu or something) about incorrect SMART readings, and the help file suggested it is nothing to worry about. This could be the case. As I told you at my question, the HDD itself has been rock solid, never acted up on me for last 23K hours.

But to verify the case, any of you guys who has a seagate HDD, preferbly 120G model, can check your SMART readings and tell me what you got?