Should I worry about Spin Retry Count?
Yousuf Khan
HD Sentinel reports its health at only 40% due to this. But it seems the
spin retries only happen during bootup. If I keep the machine running,
it never happens. It also doesn't happen during warm reboots, just
occasionally during the cold starts.
So I'm thinking it's got no real recurring problem, just a power-up
problem.
Yousuf Khan
Rod Speed
Yousuf Khan wrote:
> One of my oldest drives a 200GB Maxtor, has a spin retry count
> problem. HD Sentinel reports its health at only 40% due to this.
Post the everest SMART report.
> But it seems the spin retries only happen during bootup. If I keep the machine running, it never happens.
That's what you'd expect, because thats the only time the drive spins down.
> It also doesn't happen during warm reboots,
Because the drive doesnt spin down then.
> just occasionally during the cold starts.
It will get worse once that starts.
> So I'm thinking it's got no real recurring problem,
Fraid it is.
> just a power-up problem.
Spin retrys are only ever that.
It could be stiction.
It could be a weak power supply that cant deliver the startup current.
If you have a decent multimeter, you could see what happens
to the 12V rail during spinup. If it is a power supply problem,
you should be able to see the 12V sag during the spinup.
But if it sags, it could be stiction.
You could try another power supply if you have a spare, or
see if the drive still has a spinup problem in a different system.
Rod Speed
> Yes you should.
> Yousuf Khan wrote:
> Fraid it is.
>> just a power-up problem.
> It could be stiction.
Worth trying another power connector if it uses the older molex power connectors.
The metal tunnels can open up over time and not make very good contact and
that can produce a spinup problem at the higher than normal spin up currents.
Arno
> One of my oldest drives a 200GB Maxtor, has a spin retry count problem.
> HD Sentinel reports its health at only 40% due to this. But it seems the
> spin retries only happen during bootup. If I keep the machine running,
> it never happens. It also doesn't happen during warm reboots, just
> occasionally during the cold starts.
Not that surprisong, considering "spin retry" indicates
problems with starting the motor spindle that rotates the
platters.
> So I'm thinking it's got no real recurring problem, just a power-up
> problem.
It is a recurring problem, as it can just happen during power-up
or when you do a spin stop-start cycle. It can be that is only
happens when the disk is physically cold, as the lubricants are
a bit thicker then.
Likely fault sources: Power supply, 12V line to HDD, HDD motor
driver, lubricants getting older and thicker. Historically it could
also be stiction (heads sticking to platters), but the HDD
manufacturers claim it does not happen anymore with the surface
coatings now used.
Arno
H. P. Holm
I had 2 x 200GB DiamondMax10 drives fail last year. One refused to spin up
one morning without warning, and the other got worse on spin retry and was
unable to wake from standby or hibernate. It had the problem when testing in
3 different computers and plenty of power.
It is a very real problem and it doesn't heal itself.
--
Hans
edfair
Agree with all of the above possible external causes but anything on the
board that affects throughput of current to the motor can be a cause and
would also include the motor itself as a likely culprit.
And agree that it only gets worse until one day it doesn't spin at all.
Then everything is lost, or not economically recoverable, depending on
the value you place on the contents.
Rod Speed
> Agree with all of the above possible external causes
> but anything on the board that affects throughput of
> current to the motor can be a cause
There's fuck all of that and even less that produces
the sort of intermittent problem he is getting.
> and would also include the motor itself as a likely culprit.
Nope, that doesnt usually produce an intermittent problem.
Yousuf Khan
> Post the everest SMART report.
[ Maxtor 6L200R0 (L50148VG) ]
03 Spin Up Time 63 176 169 24587
OK: Value is normal
04 Start/Stop Count 0 253 253 1183
OK: Always passing
05 Reallocated Sector Count 63 253 253 0
OK: Value is normal
06 Read Channel Margin 100 253 253 0
OK: Value is normal
07 Seek Error Rate 0 253 252 0
OK: Always passing
08 Seek Time Performance 187 239 235 35599
OK: Value is normal
09 Power-On Time Count 0 187 187 63561
OK: Always passing
0A Spin Retry Count 157 224 206 20
OK: Value is normal
0B Calibration Retry Count 223 253 252 0
OK: Value is normal
0C Power Cycle Count 0 251 251 981
OK: Always passing
C0 Power-Off Retract Count 0 253 253 0
OK: Always passing
C1 Load/Unload Cycle Count 0 253 253 0
OK: Always passing
C2 Temperature 0 41 253 38
OK: Always passing
C3 Hardware ECC Recovered 0 253 251 21462
OK: Always passing
C4 Reallocation Event Count 0 253 253 0
OK: Always passing
C5 Current Pending Sector Count 0 253 253 0
OK: Always passing
C6 Off-Line Uncorrectable Sector Count 0 253 253 0
OK: Always passing
C7 Ultra ATA CRC Error Rate 0 199 199 0
OK: Always passing
C8 Write Error Rate 0 253 252 0
OK: Always passing
C9 <vendor-specific> 0 253 252 1
OK: Always passing
CA <vendor-specific> 0 253 252 0
OK: Always passing
CB <vendor-specific> 180 253 252 0
OK: Value is normal
CC <vendor-specific> 0 253 252 0
OK: Always passing
CD <vendor-specific> 0 253 252 0
OK: Always passing
CF <vendor-specific> 0 224 207 23
OK: Always passing
D0 <vendor-specific> 0 253 252 0
OK: Always passing
D1 <vendor-specific> 0 241 241 154
OK: Always passing
D2 <vendor-specific> 0 253 252 0
OK: Always passing
D3 <vendor-specific> 0 253 252 0
OK: Always passing
D4 <vendor-specific> 0 253 252 0
OK: Always passing
> It could be a weak power supply that cant deliver the startup current.
>
> If you have a decent multimeter, you could see what happens
> to the 12V rail during spinup. If it is a power supply problem,
> you should be able to see the 12V sag during the spinup.
>
> But if it sags, it could be stiction.
>
> You could try another power supply if you have a spare, or
> see if the drive still has a spinup problem in a different system.
Actually, the power supply was the first thing I suspected. It was
replaced a couple of months back with a relatively high-end Zalman 600W
one. The new power supply solved a bunch of other unrelated problems on
other drives in the system, but the spin retry problem still occurs on
this drive.
Yousuf Khan
Yousuf Khan
> Likely fault sources: Power supply, 12V line to HDD, HDD motor
> driver, lubricants getting older and thicker. Historically it could
> also be stiction (heads sticking to platters), but the HDD
> manufacturers claim it does not happen anymore with the surface
> coatings now used.
The power supply is new, just a few months old, and it's a 600W Zalman
now (replaced a 400W noname unit). Though other unrelated hard drive
problems have gone away after the replacement of the PS, this problem
still occurs.
Stiction sounds more like something that would've destroyed the drive
ages ago (i.e. heads touching platters sounds like it'll scratch the
platters to death). I think this spin retry problem has been occurring
for at least two years on this drive. It used to be my Windows boot
drive, but I've since moved all OS-related data away off of it, onto a
newer drive.
Yousuf Khan
Yousuf Khan
> I had 2 x 200GB DiamondMax10 drives fail last year. One refused to spin
> up one morning without warning, and the other got worse on spin retry
> and was unable to wake from standby or hibernate. It had the problem
> when testing in 3 different computers and plenty of power.
>
> It is a very real problem and it doesn't heal itself.
That's why I moved all of my OS-related data off of it, onto a new
drive. This used to be my boot drive. However, this problem has
persisted for at least two years, as far as I can tell. Yet, I've had
other drives which were both several years newer and gave much fewer
prior SMART warnings, die on me much earlier than this drive has. IOW,
I've been worried about this drive dying for several years, but instead
other drives have died ahead of it. That's why I'm getting skeptical
about the value of this spin retry warning.
Yousuf Khan
Rod Speed
> Rod Speed wrote
>> Post the everest SMART report.
> [ Maxtor 6L200R0 (L50148VG) ]
> 03 Spin Up Time 63 176 169 24587
> OK: Value is normal
Yeah, there definitely is a problem.
OK, the drive has passed its useby date. You into necrophilia ?
Obviously if you are into necrophilia, you should backup what matters
often because the usual result with spin retry problems is that one day
it just doesnt spin up at all and there is no way of getting the data back.
Rod Speed
> H. P. Holm wrote
>> I had 2 x 200GB DiamondMax10 drives fail last year. One refused to
>> spin up one morning without warning, and the other got worse on spin
>> retry and was unable to wake from standby or hibernate. It had the
>> problem when testing in 3 different computers and plenty of power.
>> It is a very real problem and it doesn't heal itself.
> That's why I moved all of my OS-related data off of it, onto a new drive. This used to be my boot drive. However, this
> problem has persisted for at least two years, as far as I can tell.
> Yet, I've had other drives which were both several years newer and gave much fewer prior SMART warnings, die on me
> much earlier than this drive has.
How many drives have you had die ? Were they all Maxtors ?
They dont like running too hot. The drive temp is currently acceptable, but that
may be because its currently in an airconditioned environment or something.
> IOW, I've been worried about this drive dying for several years, but instead other drives have died ahead of it.
> That's why I'm getting skeptical about the value of this spin retry warning.
The trouble with that line is the mechanism of spin retrys.
There isnt anything that is innocuous with spin retrys if its
not the power supply. If its the drive, it has to get worse.
H. P. Holm
> H. P. Holm wrote:
[...]
>> It is a very real problem and it doesn't heal itself.
>
> That's why I moved all of my OS-related data off of it, onto a new drive.
> This used to be my boot drive. However, this problem has persisted for at
> least two years, as far as I can tell. Yet, I've had other drives which
> were both several years newer and gave much fewer prior SMART warnings,
> die on me much earlier than this drive has. IOW, I've been worried about
> this drive dying for several years, but instead other drives have died
> ahead of it. That's why I'm getting skeptical about the value of this spin
> retry warning.
Agreed, still an acceptable and useful drive for scratch and temp files -
unpacking RAR files and sorting out stuff before storing anything important
elsewhere.
BTW: Two other 200GB DiamondMax10 from the same batch that let me down, ran
for years in an entry level Intel server 24/7 without a single hiccup until
it was replaced due to lack of storage space last October. Wonder if I
should borrow these drives and see if it will spin up now? ... Nah, life is
too short!
--
Hans
Paul
When I used to get fault it was because I was overloading the power
supply with too many additional drives (I had a couple of PCI adaptor
cards).
Rod Speed
>>> It is a very real problem and it doesn't heal itself.
>> That's why I moved all of my OS-related data off of it, onto a new
>> drive. This used to be my boot drive. However, this problem has
>> persisted for at least two years, as far as I can tell. Yet, I've
>> had other drives which were both several years newer and gave much
>> fewer prior SMART warnings, die on me much earlier than this drive
>> has. IOW, I've been worried about this drive dying for several
>> years, but instead other drives have died ahead of it. That's why
>> I'm getting skeptical about the value of this spin retry warning.
> Agreed, still an acceptable and useful drive for scratch and temp
> files - unpacking RAR files and sorting out stuff before storing
> anything important elsewhere.
The problem with that line is that modern drives are now so cheap that it makes
a hell of a lot more sense to just replace it unless you are desperately poor.
Yousuf Khan
> Yousuf Khan wrote
>> Yet, I've had other drives which were both several years newer and gave much fewer prior SMART warnings, die on me
>> much earlier than this drive has.
>
> How many drives have you had die ? Were they all Maxtors ?
> They dont like running too hot. The drive temp is currently acceptable, but that
> may be because its currently in an airconditioned environment or something.
Let me think, one was a Seagate, and others were Western Digitals. I
used to buy Maxtors all the time before, but not recently.
Yousuf Khan
Yousuf Khan
> Agreed, still an acceptable and useful drive for scratch and temp files
> - unpacking RAR files and sorting out stuff before storing anything
> important elsewhere.
Yup, I'm just using them for overflow data. Stuff I can afford to lose
and not miss.
Yousuf Khan
Yousuf Khan
> H. P. Holm wrote
>> files - unpacking RAR files and sorting out stuff before storing
>> anything important elsewhere.
>
> The problem with that line is that modern drives are now so cheap that it makes
> a hell of a lot more sense to just replace it unless you are desperately poor.
What if you're desperately cheap?
Yousuf Khan
Rod Speed
You should do the decent thing and hang yourself thoughtfully.