Support in other languages: 
Reply
Fanfold Paper
droppie
Posts: 5
Registered: ‎12-11-2009
Location: Netherlands
0

Hard drive smart high start stop count

[ Edited ]

Hello all.

 

I would like to draw the attention of the community to the some hard drive issues.

I have a SL500 2746-AFG with a Seagate ST9250827AS hard drive.

 

I used a utility called DiskCheckup to read the SMART values of the harddrive. I have googled around and found that interpretation of these values is difficult. One thing stands out however. I have a high start/stop count.

 

For smart ID #4, Start/Stop count the raw value us : 91157 after only 7 months of use.

The cooked value is 11, the worst cooked value is 11 and the threshold is 20. The utility also signals it as "FAIL". So that scared me a bit. Then i tried the seagate utility Seatools for Windows. This utility lists the smart-check as "pass" eventhough DiskCheckup lists it as "failed". I have smart enabled in the bios and i don't get any warnings at bootup.

 

So what utility should i trust? One might trust the manufactures original utility more than some third party utility. But then, this issue is not uncommon (Search google for high start/stop counts). And the raw value actually seems to be the start/stop count.  In fact, my #C1 load/unload cycle count raw value is 91278 , nearly the same.

Also i won't believe that the value reported by DisckCheckup for the cooked value and threshold is incorrect. Why would it be? So i can assume that i have a harddrive that is telling me that it is under threshold, but why does the bios not give me warning, and why does the seagate utility list it as "pass"?

 

Come to think of it, i constantly hear the clicking of the unload/load of the hard drive. And i usually work with the power plug in the outlet. And i had set windows to never shut down the harddrives when on external power a long time ago.

When googling around i also found that Seagate was not willing to answer any questions about this. They won't confirm OR deny, let alone explain and give us some more info. But i know this: constant loading and unloading IS a very bad thing mechanically. So, what to do?

 

Perhaps someone at lenovo could look into it and tell us what is going on. Maybe it's not bad at all, but then i would also like to know the answer as well as some techincal background on these smart values, just for ease of mind. Maybe Seagate will answer to a large customer like lenovo? I would appreciate some reassuring comments. The thing is, at this rate the safety margins for mechanical failiure (some say ~600000 to 1000000 counts) are uncomfortably within the economic life-span of a laptop.

 

Also, do i need to take action? Is my drive really failing?

Token Ring
Izmi
Posts: 219
Registered: ‎10-18-2008
Location: Amsterdam
0

Re: Hard drive smart high start stop count

Hi droppie,

 

What were your findings in the end?

 

I have the same type of harddisk in my T400 with a similar SMART-report and also no SMART-errors.

But the smartctl-tools report a very high start/stop-count:

 

 

Please note the following marginal Attributes:
ID# ATTRIBUTE_NAME   FLAG     VALUE WORST THRESH TYPE     UPDATED WHEN_FAILED RAW_VALUE
4 Start_Stop_Count     0x0032   001   001   020      Old_age   Always   FAILING_NOW 198823

 

 

 

What do the others think: Should the harddisk be replaced? Is this covered by warranty? If so: Can the disk be upgraded to a state-of-the-art model (possibly an SSD)?

Fanfold Paper
droppie
Posts: 5
Registered: ‎12-11-2009
Location: Netherlands

Re: Hard drive smart high start stop count

[ Edited ]

@  Izmi

I did not find any more information, but i'll share my thoughts on this anyway:

 

first of all, apparently the SMART info is not something that can be relied upon. There is no real standard and there are many differences between hard drive manufacturers and even between different types of harddrive of the same manufacturer. Any generic tool may not give an accurate picture of the hard drive health. That said, i have SMART error reporting enabled in the BIOS setup and it always reports the status of my hard drive as 'OK' at bootup, eventhough my SMART check utility thinks otherwise.

Most notably, the SEAGATE hard drive checkup utility which also does a SMART check does not report any problem. In absence of any other good reason to mistrust Seagates own hard drive checkup utilities we should probably go by what seagate gives us. In fact, if you call seagate support for an RMA they will ask you to run their utility (unless ofcourse you want to do a data recovery in which case you should not even power the drive if you think it is defective!)

 

second. I have a Seagate Momentus 5400.4 in my Thinkpad SL500. As far as i know it is built like a tank and is tailored for mobile computers. As far as I know, the constant starting and stopping with this drive is a feature of the harddrive itself and cannot be controlled with the windows settings for harddrive standby. That is why the load/unload count continues to increase eventhough you are on AC-power with harddrive standby disabled (it won't go into standby anyway due to the constant drive usage by windows). This is not much of a problem however, given that the hard drive was built with this in mind. Check the very comprehensive datasheet :

http://www.seagate.com/staticfiles/support/disc/manuals/notebook/momentus/5400.4/SATA/100468842b.pdf

It specifies the number of load/unload cycles as 600000. Yes, thats 600k cycles. Most other drives are specified at 60k cycles.

 

conclusion:

My guess is that in my case the high number of load/unload cycles is perfectly normal and that the hard drive simply does not report correct smart data (remember that the seagate check utility says it is fine). My mind was not at fully at ease though, and figuring that data is always at risk on any harddrive (and more so on mobile harddrives) i bought a network hard drive enclosure with two harddrives in raid. The stuff on my laptops harddrive that is most important is backup up to a network harddrive, which is what you should do even if your harddrive does not report a SMART failed status.

 

If my harddisk dies tomorrow, i will curse a bit, but i won't cry.

 

I am not sure if any of this applies to your case at all. I certainly recommend that you run the seagate check utility. I do believe that lenovo has a check utility on their website as well, so you should also try that. Ofcourse, if your harddrive makes strange noises or if it is very noticably slower then you should assume a bad harddrive and go into data recovery mode (and not run any utility). You could lookup the datasheet for your harddrive and find the number of unload/load cycles it can handle. This number should ofcourse not be viewed as a hard limit. The drive may fail long before or after this number of cycles.

And oh yes, backup your data often!

 

Presently my drive is at 250k cycles working mostly on AC-power. If i work on battery the number rises more quickly, but at this rate i figure it would take at least a few years before i get close to 60000 600kCycles and i hope to have replaced the laptop by then.

So no, i don't think that it should be replaced, it is not defective and thus there is no need for warranty procedure. Upgrading to SSD would help with battery life and overall speed. However, unless your operating system properly supports the TRIM command the reliability and speed of an SSD would not be much improved over magnetic hard drive.

 

But the point really is this: there are so many ways for a electronic circuit to die and there are so many modes of failiure for a mechanical harddrive that the only way to be relatively safe is to backup the data to another medium. In fact, you should even have multiple sets of backups that you can cycle to protect against an accidental overwrite of your backup with bad data (eg due to plain stupidity during a data recovery scenario or things like virusses)

 

Hope this helps. Please don't take any of this to be the best advice possible. If you really are concerned about your data then you should get an official statement on this by lenovo/seagate. But better still, make backups often.

 

edit: changed 60000 to 600kCycles . one zero. huge difference :smileyhappy: