* Re: Use of WD20EARS with MDADM
2010-04-21 13:20 ` Bill Davidsen
@ 2010-04-21 13:45 ` Tim Small
2010-04-21 14:32 ` Mikael Abrahamsson
` (3 subsequent siblings)
4 siblings, 0 replies; 52+ messages in thread
From: Tim Small @ 2010-04-21 13:45 UTC (permalink / raw)
To: Bill Davidsen; +Cc: Phillip Susi, st0ff, linux-raid
Bill Davidsen wrote:
>> I believe that these disks only come in the "green" variety. I recently
>> picked up a 1.5 tb version for testing and cheap bulk storage, and I
>> would not suggest using them in a raid array because the green drives
>> firmware automatically parks the head after 8 seconds of inactivity and
>> reduces the rpm of the disk. The constant parking can quickly wear out
>> the head under high use and there is no way to disable this "feature".
>>
>
> I hear this said, but I don't have any data to back it up. Drive
> vendors aren't stupid, so if the parking feature is likely to cause
> premature failures under warranty, I would expect that the feature
> would not be there, or that the drive would be made more robust. Maybe
> I have too much faith in greed as a design goal, but I have to wonder
> if load cycles are as destructive as seems to be the assumption.
I've used the 500G, and 2TB WD consumer green drives in md arrays, not
many - about 10 in total. The older 500G drives (like many WD 2.5"
drives) did frequent head unload/reloads under Linux due to interactions
with the default timings of the Linux block layer, I believe. This can
be fixed using WD's wdidle3.exe under DOS. You can monitor this using
smartctl - look at the raw value for unloads, but it shouldn't be an
issue with newer drives.
Both the 2TB and 500G drives seem to lock up and need resetting by Linux
(this happens automatically) if you poll the SMART status e.g. you run
smartd, or munin+smartctl etc. (the 2TB drives that I've had also seem
to occasionally lock up under other workloads, but you might be able to
live with this) - the SMART thing might be a firmware bug but WDC say
"contact your Linux vendor" - they only provide support for Windows - I
must get around to polling one of them for SMART under Windows XP for
comparison.
If you're going to be trying to use a load of them in a RAID, then you
need to be careful about vibration damping - they may not cope that well
with vibration (check SMART high-fly-writes and hardware-ecc-recovered
raw values if available). They are also not rated for 24/7 operation, I
believe.
Tim.
--
South East Open Source Solutions Limited
Registered in England and Wales with company number 06134732.
Registered Office: 2 Powell Gardens, Redhill, Surrey, RH1 1TQ
VAT number: 900 6633 53 http://seoss.co.uk/ +44-(0)1273-808309
^ permalink raw reply [flat|nested] 52+ messages in thread* Re: Use of WD20EARS with MDADM
2010-04-21 13:20 ` Bill Davidsen
2010-04-21 13:45 ` Tim Small
@ 2010-04-21 14:32 ` Mikael Abrahamsson
2010-04-21 21:58 ` Berkey B Walker
2010-04-21 15:14 ` Phillip Susi
` (2 subsequent siblings)
4 siblings, 1 reply; 52+ messages in thread
From: Mikael Abrahamsson @ 2010-04-21 14:32 UTC (permalink / raw)
To: Bill Davidsen; +Cc: Phillip Susi, st0ff, linux-raid
On Wed, 21 Apr 2010, Bill Davidsen wrote:
> I hear this said, but I don't have any data to back it up. Drive vendors
> aren't stupid, so if the parking feature is likely to cause premature
> failures under warranty, I would expect that the feature would not be there,
> or that the drive would be made more robust. Maybe I have too much faith in
> greed as a design goal, but I have to wonder if load cycles are as
> destructive as seems to be the assumption.
What I think people are worried about is that a drive might have X
load/unload cycles in the data sheet (300k or 600k seem to be normal
figures) and reaching this in 1-2 years of "normal" (according to the
user who is running it 24/7) might be worrying (and understandably so).
Otoh these drives seem to be designed for desktop 8 hour per day use, so
running them as a 24/7 fileserver under linux is not what they were
designed for. I have no idea what will happen when the load/unload cycles
goes over the data sheet number, but my guess is that it was put there for
a reason.
> I'd love to find some real data, anecdotal stories about older drives are not
> overly helpful. Clearly there is a trade-off between energy saving, response,
> and durability, I just don't have any data from a large population of new
> (green) drives.
My personal experience from the WD20EADS drives is that around 40% of
them failed within the first year of operation. This is not from a large
population of drives though and wasn't due to load/unload cycles. I had no
problem getting them replaced under warranty, but I'm running RAID6
nowadays :P
--
Mikael Abrahamsson email: swmike@swm.pp.se
^ permalink raw reply [flat|nested] 52+ messages in thread* Re: Use of WD20EARS with MDADM
2010-04-21 14:32 ` Mikael Abrahamsson
@ 2010-04-21 21:58 ` Berkey B Walker
2010-05-02 22:33 ` Bill Davidsen
0 siblings, 1 reply; 52+ messages in thread
From: Berkey B Walker @ 2010-04-21 21:58 UTC (permalink / raw)
To: Mikael Abrahamsson; +Cc: Bill Davidsen, Phillip Susi, st0ff, linux-raid
Mikael Abrahamsson wrote:
> On Wed, 21 Apr 2010, Bill Davidsen wrote:
>
>> I hear this said, but I don't have any data to back it up. Drive
>> vendors aren't stupid, so if the parking feature is likely to cause
>> premature failures under warranty, I would expect that the feature
>> would not be there, or that the drive would be made more robust.
>> Maybe I have too much faith in greed as a design goal, but I have to
>> wonder if load cycles are as destructive as seems to be the assumption.
>
> What I think people are worried about is that a drive might have X
> load/unload cycles in the data sheet (300k or 600k seem to be normal
> figures) and reaching this in 1-2 years of "normal" (according to the
> user who is running it 24/7) might be worrying (and understandably so).
>
> Otoh these drives seem to be designed for desktop 8 hour per day use,
> so running them as a 24/7 fileserver under linux is not what they were
> designed for. I have no idea what will happen when the load/unload
> cycles goes over the data sheet number, but my guess is that it was
> put there for a reason.
>
>> I'd love to find some real data, anecdotal stories about older drives
>> are not overly helpful. Clearly there is a trade-off between energy
>> saving, response, and durability, I just don't have any data from a
>> large population of new (green) drives.
>
> My personal experience from the WD20EADS drives is that around 40% of
> them failed within the first year of operation. This is not from a
> large population of drives though and wasn't due to load/unload
> cycles. I had no problem getting them replaced under warranty, but I'm
> running RAID6 nowadays :P
>
Sorry, you sound like a factory droid. *I* see no reason for early
failure besides cheap mat'ls in construction. Were these assertations
of short life to be true, I would campaign against the drive maker. (I
think that they are just normalizing failure rate against warranty
claims) Buy good stuff. I *wish* I could define the term by mfg. It
seems Seagate, & WD don't hack it. The Japanese drives did, but since
the $ dropped -
One thing seemingly missed is the relationship between storage density
and drive temp.variations.. Hard drive mfgs are going to be in deep
doodoo when the SSD folks get price/perf in the lead lane. This year, I
predict. And maybe another 2 for long term reliability to be in the lead..
I believe that many [most?] RAID users are looking for results (long
term archival) that are not intended in the design.We are about 2
generations away from that being a reality - I think.. For other users,
I would suggest a mirror machine. with both machines being scrubbed
daily, and media being dissimilar in mfg and mfg date.
I can't wait until Neil gets to (has to) play/work with the coming tech.
Neat things are coming.
b-
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 21:58 ` Berkey B Walker
@ 2010-05-02 22:33 ` Bill Davidsen
2010-05-03 0:08 ` Berkey B Walker
0 siblings, 1 reply; 52+ messages in thread
From: Bill Davidsen @ 2010-05-02 22:33 UTC (permalink / raw)
To: Berkey B Walker; +Cc: Mikael Abrahamsson, Phillip Susi, st0ff, linux-raid
Berkey B Walker wrote:
>
>
> Mikael Abrahamsson wrote:
>> On Wed, 21 Apr 2010, Bill Davidsen wrote:
>>
>>> I hear this said, but I don't have any data to back it up. Drive
>>> vendors aren't stupid, so if the parking feature is likely to cause
>>> premature failures under warranty, I would expect that the feature
>>> would not be there, or that the drive would be made more robust.
>>> Maybe I have too much faith in greed as a design goal, but I have to
>>> wonder if load cycles are as destructive as seems to be the assumption.
>>
>> What I think people are worried about is that a drive might have X
>> load/unload cycles in the data sheet (300k or 600k seem to be normal
>> figures) and reaching this in 1-2 years of "normal" (according to the
>> user who is running it 24/7) might be worrying (and understandably so).
>>
>> Otoh these drives seem to be designed for desktop 8 hour per day use,
>> so running them as a 24/7 fileserver under linux is not what they
>> were designed for. I have no idea what will happen when the
>> load/unload cycles goes over the data sheet number, but my guess is
>> that it was put there for a reason.
>>
>>> I'd love to find some real data, anecdotal stories about older
>>> drives are not overly helpful. Clearly there is a trade-off between
>>> energy saving, response, and durability, I just don't have any data
>>> from a large population of new (green) drives.
>>
>> My personal experience from the WD20EADS drives is that around 40% of
>> them failed within the first year of operation. This is not from a
>> large population of drives though and wasn't due to load/unload
>> cycles. I had no problem getting them replaced under warranty, but
>> I'm running RAID6 nowadays :P
>>
> Sorry, you sound like a factory droid. *I* see no reason for early
> failure besides cheap mat'ls in construction. Were these assertations
> of short life to be true, I would campaign against the drive maker.
> (I think that they are just normalizing failure rate against warranty
> claims) Buy good stuff. I *wish* I could define the term by mfg. It
> seems Seagate, & WD don't hack it. The Japanese drives did, but since
> the $ dropped -
>
Let's see, first you put my name on something I was quoting (with
attribution), delete the correct name of the person you are quoting, and
then call me a "factory droid." So I have some idea of your attention to
detail. Second, the short term failure rates are influenced by
components delivered, assembly, and treatment in shipping. So assembly
is controlled by the vendor, parts are influenced by supplier selected,
and delivery treatment is usually selected by the retailer. A local
clone maker found that delicate parts delivered on Wednesday had a
higher infant mortality that other days. Regular driver had Wednesday
off, sub thought "drop ship" was an unloading method, perhaps.
> One thing seemingly missed is the relationship between storage density
> and drive temp.variations.. Hard drive mfgs are going to be in deep
> doodoo when the SSD folks get price/perf in the lead lane. This year,
> I predict. And maybe another 2 for long term reliability to be in the
> lead..
>
I think you're an optimist on cost equality, people are changing to
green drives which are generally slower due to spin down or lower rpm,
because the cost of power and cooling is important. It's not clear if
current SSD tech will be around in five years, because there are new
technologies coming which are inherently far more stable for multiple
writes. The spinning platter may be ending, but the replacement is not
in sight. In ten years I doubt current SSD tech will be in use, replaced
by phase change, optical isomers, electron spin, or something still in a
lab. And the deployment of user visible large sectors (write chunks,
whatever) is not clear, if the next tech will work just as well with
smaller sectors, this may become a moot point.
> I believe that many [most?] RAID users are looking for results (long
> term archival) that are not intended in the design.We are about 2
> generations away from that being a reality - I think.. For other
> users, I would suggest a mirror machine. with both machines being
> scrubbed daily, and media being dissimilar in mfg and mfg date.
>
It's not clear that date of manufacture is particularly critical, while
date of deployment (in-use hours), probably is. But looking at the
Google disk paper, a crate of drives from the same batch doesn't all
drop dead at once, or close to it, so age in service is a factor, but
not likely a critical factor.
> I can't wait until Neil gets to (has to) play/work with the coming
> tech. Neat things are coming.
>
I would rather see some of the many things on the "someday list" get
implemented. It's more fun to play with new stuff than polish off the
uglies in the old, but the uglies are still there.
--
Bill Davidsen <davidsen@tmr.com>
"We can't solve today's problems by using the same thinking we
used in creating them." - Einstein
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-05-02 22:33 ` Bill Davidsen
@ 2010-05-03 0:08 ` Berkey B Walker
0 siblings, 0 replies; 52+ messages in thread
From: Berkey B Walker @ 2010-05-03 0:08 UTC (permalink / raw)
To: Bill Davidsen; +Cc: Mikael Abrahamsson, Phillip Susi, st0ff, linux-raid
Bill Davidsen wrote:
> Berkey B Walker wrote:
>> Snip...
> Let's see, first you put my name on something I was quoting (with
> attribution), delete the correct name of the person you are quoting,
> and then call me a "factory droid." So I have some idea of your
> attention to detail.
..snip
> I would rather see some of the many things on the "someday list" get
> implemented. It's more fun to play with new stuff than polish off the
> uglies in the old, but the uglies are still there.
>
Items snipped to reduce net bandwidth.
Sir, I have had and have a high regard for your posts. I will gladly
give you a pass for having a seemingly unpleasant weekend. I have gone
through MY archives in this manner and have found absolutely NO
discrepancies between my copy of the posting to which I replied (which
was NOT addressed to you, but was a complete quote, my "sent" post, the
board's posting of that post, or..... your post to me (board copied).
I should probably just take heat in general, for some of my postings
back in my drinking days.
Sorry to have been a part of whatever was going on.
Best to you,
b-
All of the quotes are properly done. The post (read carefully) was
addresses to Mikael Abrahamsson
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 13:20 ` Bill Davidsen
2010-04-21 13:45 ` Tim Small
2010-04-21 14:32 ` Mikael Abrahamsson
@ 2010-04-21 15:14 ` Phillip Susi
2010-04-21 16:42 ` Bill Davidsen
2010-04-21 17:36 ` Mark Knecht
2010-04-21 15:52 ` Simon Matthews
2010-04-21 19:24 ` Richard Scobie
4 siblings, 2 replies; 52+ messages in thread
From: Phillip Susi @ 2010-04-21 15:14 UTC (permalink / raw)
To: Bill Davidsen; +Cc: st0ff, linux-raid
On 4/21/2010 9:20 AM, Bill Davidsen wrote:
> I hear this said, but I don't have any data to back it up. Drive vendors
> aren't stupid, so if the parking feature is likely to cause premature
> failures under warranty, I would expect that the feature would not be
> there, or that the drive would be made more robust. Maybe I have too
> much faith in greed as a design goal, but I have to wonder if load
> cycles are as destructive as seems to be the assumption.
Indeed, I think you have too much faith in people doing sensible things.
Especially when their average customer isn't placing the drive in a
high use environment and they know it, and suggest against doing so.
> I'd love to find some real data, anecdotal stories about older drives
> are not overly helpful. Clearly there is a trade-off between energy
> saving, response, and durability, I just don't have any data from a
> large population of new (green) drives.
I've not seen any anecdotal stories, but I have seen plenty of reports
with real data showing a large number of head unloads from the SMART
data after a relatively short period of use. Personally mine has a few
hundred so far and I have not even used it for real storage yet, only
testing. The specifications say it's good for 300,000 cycles, so do the
math... getting 5 unloads per minute would lead to probable failure
after 41 days. Granted that is about worst case, but still something to
watch out for. In order to make it the entire 3 year warranty period,
you need to stay under 11.4 unloads per hour. If you have very little
IO activity, or VERY MUCH, then this is entirely possible, but more
moderate loads in the middle have been observed to cause hundreds of
unloads per hour.
Given that, and the fact that WD themselves have stated that you should
not use these drives in a raid array, I'd either stay away, or watch out
for this problem and try to take action to avoid and monitor it.
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 15:14 ` Phillip Susi
@ 2010-04-21 16:42 ` Bill Davidsen
2010-04-21 17:36 ` Mark Knecht
1 sibling, 0 replies; 52+ messages in thread
From: Bill Davidsen @ 2010-04-21 16:42 UTC (permalink / raw)
To: Phillip Susi; +Cc: st0ff, linux-raid
Phillip Susi wrote:
> On 4/21/2010 9:20 AM, Bill Davidsen wrote:
>
>> I hear this said, but I don't have any data to back it up. Drive vendors
>> aren't stupid, so if the parking feature is likely to cause premature
>> failures under warranty, I would expect that the feature would not be
>> there, or that the drive would be made more robust. Maybe I have too
>> much faith in greed as a design goal, but I have to wonder if load
>> cycles are as destructive as seems to be the assumption.
>>
>
> Indeed, I think you have too much faith in people doing sensible things.
> Especially when their average customer isn't placing the drive in a
> high use environment and they know it, and suggest against doing so.
>
>
>> I'd love to find some real data, anecdotal stories about older drives
>> are not overly helpful. Clearly there is a trade-off between energy
>> saving, response, and durability, I just don't have any data from a
>> large population of new (green) drives.
>>
>
> I've not seen any anecdotal stories, but I have seen plenty of reports
> with real data showing a large number of head unloads from the SMART
> data after a relatively short period of use. Personally mine has a few
> hundred so far and I have not even used it for real storage yet, only
> testing. The specifications say it's good for 300,000 cycles, so do the
> math... getting 5 unloads per minute would lead to probable failure
> after 41 days. Granted that is about worst case, but still something to
> watch out for. In order to make it the entire 3 year warranty period,
> you need to stay under 11.4 unloads per hour. If you have very little
> IO activity, or VERY MUCH, then this is entirely possible, but more
> moderate loads in the middle have been observed to cause hundreds of
> unloads per hour.
>
> Given that, and the fact that WD themselves have stated that you should
> not use these drives in a raid array, I'd either stay away, or watch out
> for this problem and try to take action to avoid and monitor it.
>
Part of this is my feeling that no one really knows if the drive fails
after N loads, because even if WD could set the unload time down, the
cycle takes time to happen, so I would bet that they are taking an
educated guess. The other part is that there are lots of clerical tasks
which would hit the drive, under Windows, single drive, 3-5 times a
minute. Data entry comes to mind, customer support, print servers, etc.
Granted that these are probably 7x5 hours a week, but I'm think 2/min,
7hr/day, 200day/yr... 168k/yr, and that's not the worst case.
Having run lots of drives (some TB of 73GB 15k rpm LVD320) MTTF is
interesting, because the curve has spikes at the front from infant
mortality, and at the end from old age, but it was damn quiet in the
middle. I'd love to see the data on these, not because I'm going to run
them but just to keep current, so when someone calls me and says they
got a great deal on green drives, I'll know what to tell them.
--
Bill Davidsen <davidsen@tmr.com>
"We can't solve today's problems by using the same thinking we
used in creating them." - Einstein
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 15:14 ` Phillip Susi
2010-04-21 16:42 ` Bill Davidsen
@ 2010-04-21 17:36 ` Mark Knecht
2010-04-21 18:40 ` Tim Small
1 sibling, 1 reply; 52+ messages in thread
From: Mark Knecht @ 2010-04-21 17:36 UTC (permalink / raw)
To: Phillip Susi; +Cc: Bill Davidsen, st0ff, linux-raid
On Wed, Apr 21, 2010 at 8:14 AM, Phillip Susi <psusi@cfl.rr.com> wrote:
<SNIP>
> I've not seen any anecdotal stories, but I have seen plenty of reports
> with real data showing a large number of head unloads from the SMART
> data after a relatively short period of use. Personally mine has a few
> hundred so far and I have not even used it for real storage yet, only
> testing. The specifications say it's good for 300,000 cycles, so do the
> math... getting 5 unloads per minute would lead to probable failure
> after 41 days. Granted that is about worst case, but still something to
> watch out for. In order to make it the entire 3 year warranty period,
> you need to stay under 11.4 unloads per hour. If you have very little
> IO activity, or VERY MUCH, then this is entirely possible, but more
> moderate loads in the middle have been observed to cause hundreds of
> unloads per hour.
>
> Given that, and the fact that WD themselves have stated that you should
> not use these drives in a raid array, I'd either stay away, or watch out
> for this problem and try to take action to avoid and monitor it.
I think I reported this earlier, but here is a WD10EARS drive in a
standard Gentoo Linux desktop machine. The drive has 1661 hours
powered up and 43508 load cycles. That's 26/hour and works out to
about 14 months before it will be out of spec at 300K cycles.
This machine only gets a few hours of use a day but is generally
powered up all the time. I don't know why Linux wakes this drive up
roughly every two minutes, assuming it's Linux and not the drive
itself or something on the motherboard, but it does.
I tried three of these drives in a RAID1 in another machine and they
simply didn't work. Went off line over and over, with big wait times
when they were working. I made no other changes expect changing to a
WD 500GB RAID Edition drive and get almostno load cycle counts at all.
- Mark
gandalf ~ # smartctl -i /dev/sda
smartctl 5.39.1 2010-01-28 r3054 [x86_64-pc-linux-gnu] (local build)
Copyright (C) 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net
=== START OF INFORMATION SECTION ===
Device Model: WDC WD10EARS-00Y5B1
Serial Number: WD-WCAV55464493
Firmware Version: 80.00A80
User Capacity: 1,000,204,886,016 bytes
Device is: Not in smartctl database [for details use: -P showall]
ATA Version is: 8
ATA Standard is: Exact ATA specification draft version not indicated
Local Time is: Wed Apr 21 10:35:32 2010 PDT
SMART support is: Available - device has SMART capability.
SMART support is: Enabled
gandalf ~ #
gandalf ~ # smartctl -A /dev/sda
smartctl 5.39.1 2010-01-28 r3054 [x86_64-pc-linux-gnu] (local build)
Copyright (C) 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net
=== START OF READ SMART DATA SECTION ===
SMART Attributes Data Structure revision number: 16
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE
UPDATED WHEN_FAILED RAW_VALUE
1 Raw_Read_Error_Rate 0x002f 200 200 051 Pre-fail
Always - 0
3 Spin_Up_Time 0x0027 129 128 021 Pre-fail
Always - 6525
4 Start_Stop_Count 0x0032 100 100 000 Old_age
Always - 21
5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail
Always - 0
7 Seek_Error_Rate 0x002e 200 200 000 Old_age
Always - 0
9 Power_On_Hours 0x0032 098 098 000 Old_age
Always - 1661
10 Spin_Retry_Count 0x0032 100 253 000 Old_age
Always - 0
11 Calibration_Retry_Count 0x0032 100 253 000 Old_age
Always - 0
12 Power_Cycle_Count 0x0032 100 100 000 Old_age
Always - 20
192 Power-Off_Retract_Count 0x0032 200 200 000 Old_age
Always - 5
193 Load_Cycle_Count 0x0032 186 186 000 Old_age
Always - 43508
194 Temperature_Celsius 0x0022 121 116 000 Old_age
Always - 26
196 Reallocated_Event_Count 0x0032 200 200 000 Old_age
Always - 0
197 Current_Pending_Sector 0x0032 200 200 000 Old_age
Always - 0
198 Offline_Uncorrectable 0x0030 200 200 000 Old_age
Offline - 0
199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age
Always - 0
200 Multi_Zone_Error_Rate 0x0008 200 200 000 Old_age
Offline - 0
gandalf ~ #
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 17:36 ` Mark Knecht
@ 2010-04-21 18:40 ` Tim Small
2010-04-21 19:01 ` Mark Knecht
2010-04-22 16:13 ` Use of WD20EARS with MDADM Khelben Blackstaff
0 siblings, 2 replies; 52+ messages in thread
From: Tim Small @ 2010-04-21 18:40 UTC (permalink / raw)
To: Mark Knecht; +Cc: Phillip Susi, Bill Davidsen, st0ff, linux-raid
Mark Knecht wrote:
> This machine only gets a few hours of use a day but is generally
> powered up all the time. I don't know why Linux wakes this drive up
> roughly every two minutes, assuming it's Linux and not the drive
> itself or something on the motherboard, but it does.
>
This is documented here:
https://ata.wiki.kernel.org/index.php/Known_issues#Drives_which_perform_frequent_head_unloads_under_Linux
If you want to fix it, then wdidle3.exe worked for me. Search for:
wdidle3_1_00.zip
Tim.
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 18:40 ` Tim Small
@ 2010-04-21 19:01 ` Mark Knecht
2010-04-21 19:31 ` Clinton Lee Taylor
` (2 more replies)
2010-04-22 16:13 ` Use of WD20EARS with MDADM Khelben Blackstaff
1 sibling, 3 replies; 52+ messages in thread
From: Mark Knecht @ 2010-04-21 19:01 UTC (permalink / raw)
To: Tim Small; +Cc: Phillip Susi, Bill Davidsen, st0ff, linux-raid
On Wed, Apr 21, 2010 at 11:40 AM, Tim Small <tim@buttersideup.com> wrote:
> Mark Knecht wrote:
>> This machine only gets a few hours of use a day but is generally
>> powered up all the time. I don't know why Linux wakes this drive up
>> roughly every two minutes, assuming it's Linux and not the drive
>> itself or something on the motherboard, but it does.
>>
>
>
> This is documented here:
>
> https://ata.wiki.kernel.org/index.php/Known_issues#Drives_which_perform_frequent_head_unloads_under_Linux
>
> If you want to fix it, then wdidle3.exe worked for me. Search for:
>
> wdidle3_1_00.zip
>
>
> Tim.
>
Yeah, that's been reported here before, but how does someone run this
Windows program on a remote machine that boots only Linux? Even if it
was a DOS executable the machine has no floppy. I presume you are
dual-boot with Windows? If so maybe I'll install Windows the next time
I visit.
I presume I cannot run this in a VM?
It's unfortunate that this drive doesn't respond to APM commands either.
- Mark
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 52+ messages in thread* Re: Use of WD20EARS with MDADM
2010-04-21 19:01 ` Mark Knecht
@ 2010-04-21 19:31 ` Clinton Lee Taylor
2010-04-22 0:51 ` Steven Haigh
2010-04-21 19:33 ` Phillip Susi
2010-04-22 11:40 ` wdidle3 Tim Small
2 siblings, 1 reply; 52+ messages in thread
From: Clinton Lee Taylor @ 2010-04-21 19:31 UTC (permalink / raw)
To: Mark Knecht; +Cc: Tim Small, Phillip Susi, Bill Davidsen, st0ff, linux-raid
Greetings ...
On 21 April 2010 21:01, Mark Knecht <markknecht@gmail.com> wrote:
> On Wed, Apr 21, 2010 at 11:40 AM, Tim Small <tim@buttersideup.com> wrote:
>> Mark Knecht wrote:
>>> This machine only gets a few hours of use a day but is generally
>>> powered up all the time. I don't know why Linux wakes this drive up
>>> roughly every two minutes, assuming it's Linux and not the drive
>>> itself or something on the motherboard, but it does.
>>>
>>
>>
>> This is documented here:
>>
>> https://ata.wiki.kernel.org/index.php/Known_issues#Drives_which_perform_frequent_head_unloads_under_Linux
>>
>> If you want to fix it, then wdidle3.exe worked for me. Search for:
>>
>> wdidle3_1_00.zip
>>
>>
>> Tim.
>>
>
> Yeah, that's been reported here before, but how does someone run this
> Windows program on a remote machine that boots only Linux? Even if it
> was a DOS executable the machine has no floppy. I presume you are
> dual-boot with Windows? If so maybe I'll install Windows the next time
> I visit.
Why don't you download FreeDOS boot floppy image. Add the DOS
program to the image. Copy image to your boot partition, add an entry
in your Boot Manager to boot the image using some line syslinux image
booter.
On CentOS, install syslinux package using "yum install syslinux"
Copy memdisk to your boot partition
From the www.freedos.org web site, download floppy image in
http://www.freedos.org/freedos/files/ ...
wget -vb http://www.ibiblio.org/pub/micro/pc-stuff/freedos/files/distributions/1.0/fdboot.img
Mount the downloaded image file with "mount fdboot.img /mnt/floppy/ -o rw,loop"
Copy DOS program into the floppy image.
Add to your grub.conf file ...
title FreeDOS - Boot Image
kernel memdisk
initrd fdboot.img
Now reboot your system that does not have a floppy drive off the
FreeDOS image and run the DOS program to make changes. The only thing
I can think of that might be a problem, is that you might have the
drive attached to a SATA port that is not supported by the DOS program
or something like that.
This is what I use to do BIOS and other firmware updates, that Linux
does not have a tool for.
Mailed
LeeT
P.S. Sorry for the Off Topic post, but I hope it helps.
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 52+ messages in thread* Re: Use of WD20EARS with MDADM
2010-04-21 19:31 ` Clinton Lee Taylor
@ 2010-04-22 0:51 ` Steven Haigh
0 siblings, 0 replies; 52+ messages in thread
From: Steven Haigh @ 2010-04-22 0:51 UTC (permalink / raw)
To: Clinton Lee Taylor; +Cc: linux-raid
+1 Insightful.
This is a gem of a tip - so much so that I've archived the text of this as
I KNOW it will come in handy soon.
It may be been semi-off topic, but it is worth its weight in gold!
Thankyou!
On Wed, 21 Apr 2010 21:31:27 +0200, Clinton Lee Taylor
<clintonlee.taylor@gmail.com> wrote:
> Greetings ...
>
> On 21 April 2010 21:01, Mark Knecht <markknecht@gmail.com> wrote:
>> On Wed, Apr 21, 2010 at 11:40 AM, Tim Small <tim@buttersideup.com>
wrote:
>>> Mark Knecht wrote:
>>>> This machine only gets a few hours of use a day but is generally
>>>> powered up all the time. I don't know why Linux wakes this drive up
>>>> roughly every two minutes, assuming it's Linux and not the drive
>>>> itself or something on the motherboard, but it does.
>>>>
>>>
>>>
>>> This is documented here:
>>>
>>>
https://ata.wiki.kernel.org/index.php/Known_issues#Drives_which_perform_frequent_head_unloads_under_Linux
>>>
>>> If you want to fix it, then wdidle3.exe worked for me. Search for:
>>>
>>> wdidle3_1_00.zip
>>>
>>>
>>> Tim.
>>>
>>
>> Yeah, that's been reported here before, but how does someone run this
>> Windows program on a remote machine that boots only Linux? Even if it
>> was a DOS executable the machine has no floppy. I presume you are
>> dual-boot with Windows? If so maybe I'll install Windows the next time
>> I visit.
> Why don't you download FreeDOS boot floppy image. Add the DOS
> program to the image. Copy image to your boot partition, add an entry
> in your Boot Manager to boot the image using some line syslinux image
> booter.
>
> On CentOS, install syslinux package using "yum install syslinux"
> Copy memdisk to your boot partition
>
> From the www.freedos.org web site, download floppy image in
> http://www.freedos.org/freedos/files/ ...
>
> wget -vb
>
http://www.ibiblio.org/pub/micro/pc-stuff/freedos/files/distributions/1.0/fdboot.img
>
> Mount the downloaded image file with "mount fdboot.img /mnt/floppy/ -o
> rw,loop"
>
> Copy DOS program into the floppy image.
>
> Add to your grub.conf file ...
>
> title FreeDOS - Boot Image
> kernel memdisk
> initrd fdboot.img
>
> Now reboot your system that does not have a floppy drive off the
> FreeDOS image and run the DOS program to make changes. The only thing
> I can think of that might be a problem, is that you might have the
> drive attached to a SATA port that is not supported by the DOS program
> or something like that.
>
> This is what I use to do BIOS and other firmware updates, that Linux
> does not have a tool for.
>
> Mailed
> LeeT
>
> P.S. Sorry for the Off Topic post, but I hope it helps.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-raid" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
Steven Haigh
Email: netwiz@crc.id.au
Web: http://www.crc.id.au
Phone: (03) 9001 6090 - 0412 935 897
Fax: (03) 8338 0299
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 19:01 ` Mark Knecht
2010-04-21 19:31 ` Clinton Lee Taylor
@ 2010-04-21 19:33 ` Phillip Susi
2010-04-21 20:36 ` Mark Knecht
2010-05-12 13:06 ` Tim Small
2010-04-22 11:40 ` wdidle3 Tim Small
2 siblings, 2 replies; 52+ messages in thread
From: Phillip Susi @ 2010-04-21 19:33 UTC (permalink / raw)
To: Mark Knecht; +Cc: Tim Small, Bill Davidsen, st0ff, linux-raid
On 4/21/2010 3:01 PM, Mark Knecht wrote:
> It's unfortunate that this drive doesn't respond to APM commands either.
Yes it is.... would be nice if they made it respond to the standard APM
command meant to configure this kind of behavior instead of creating a
proprietary command and dos utility to invoke it. It's also a shame the
drive lies about its physical sector size and has no way to turn that
off. Might be a nice project to reverse engineer this utility on
windows to figure out the command it sends down and add it to hdparm.
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 19:33 ` Phillip Susi
@ 2010-04-21 20:36 ` Mark Knecht
2010-05-12 13:06 ` Tim Small
1 sibling, 0 replies; 52+ messages in thread
From: Mark Knecht @ 2010-04-21 20:36 UTC (permalink / raw)
To: Phillip Susi; +Cc: Tim Small, Bill Davidsen, st0ff, linux-raid
On Wed, Apr 21, 2010 at 12:33 PM, Phillip Susi <psusi@cfl.rr.com> wrote:
> On 4/21/2010 3:01 PM, Mark Knecht wrote:
>> It's unfortunate that this drive doesn't respond to APM commands either.
>
> Yes it is.... would be nice if they made it respond to the standard APM
> command meant to configure this kind of behavior instead of creating a
> proprietary command and dos utility to invoke it. It's also a shame the
> drive lies about its physical sector size and has no way to turn that
> off. Might be a nice project to reverse engineer this utility on
> windows to figure out the command it sends down and add it to hdparm.
I thought the same thing so I worked with the hdparm developer (Mark
Lord) and gave him data two months back. There was a command he had me
run (hdparm --istdout /dev/sda) that extracts firmware tables or
something and he spent time looking at the tables only to decide that
WD simply isn't identifying anywhere in the tables that the drive is
4K/ physical sector. With that it doesn't seem that hdparm could ever
do anything automatic that would be safe.
- Mark
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 19:33 ` Phillip Susi
2010-04-21 20:36 ` Mark Knecht
@ 2010-05-12 13:06 ` Tim Small
1 sibling, 0 replies; 52+ messages in thread
From: Tim Small @ 2010-05-12 13:06 UTC (permalink / raw)
To: Phillip Susi; +Cc: Mark Knecht, Bill Davidsen, st0ff, linux-raid
On 21/04/10 20:33, Phillip Susi wrote:
> On 4/21/2010 3:01 PM, Mark Knecht wrote:
>
>> It's unfortunate that this drive doesn't respond to APM commands either.
>>
> Yes it is.... would be nice if they made it respond to the standard APM
> command meant to configure this kind of behavior instead of creating a
> proprietary command and dos utility to invoke it. It's also a shame the
> drive lies about its physical sector size and has no way to turn that
> off. Might be a nice project to reverse engineer this utility on
> windows to figure out the command it sends down and add it to hdparm.
>
I did have a look at this a while ago (it seemed to be built with Watcom
(now open-source) and a public-domain DOS extended along with an
open-sourced binary-compressor - decompilation under Watcom on FreeDOS
seemed possible), and it seems that Mark Lord has done a chunk more
hacking on it, since there is a reference to initial support in the
latest hdparm changelog - which you might want to take a look at....
Cheers,
Tim.
^ permalink raw reply [flat|nested] 52+ messages in thread
* wdidle3
2010-04-21 19:01 ` Mark Knecht
2010-04-21 19:31 ` Clinton Lee Taylor
2010-04-21 19:33 ` Phillip Susi
@ 2010-04-22 11:40 ` Tim Small
2 siblings, 0 replies; 52+ messages in thread
From: Tim Small @ 2010-04-22 11:40 UTC (permalink / raw)
To: Mark Knecht; +Cc: Phillip Susi, Bill Davidsen, st0ff, linux-raid
Mark Knecht wrote:
>> [wdidle3.exe]...
> Yeah, that's been reported here before, but how does someone run this
> Windows program on a remote machine that boots only Linux? Even if it
> was a DOS executable the machine has no floppy.
It is a DOS executable. As previously mentioned, you can use:
USB floppy emulation (if the BIOS supports it)
USB HD emulation
Booting from an HD partition (e.g. temporarily break RAID for one of the
swap partitons, and install DOS in that partition).
syslinux + memdisk + image file
pxelinux + memdisk + image file
I use the latter for using DOS bios upgrades etc. remotely using IPMI+SoL.
If anyone want to try and reverse-engineer wdidle3, it appears to have
been compiled with Watcom (now open source), and a free DOS extender and
then packed (again using an open tool). You can run it under the Watcom
disassembler / debugger (which is packaged with FreeDOS).
With all this open source software rattling around Western Digital, it'd
be nice if they just did a bloody Linux version (or at least documented
the vendor-specific commands which they use).
Tim.
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 18:40 ` Tim Small
2010-04-21 19:01 ` Mark Knecht
@ 2010-04-22 16:13 ` Khelben Blackstaff
2010-04-22 18:16 ` Simon Matthews
2010-04-22 19:44 ` Phillip Susi
1 sibling, 2 replies; 52+ messages in thread
From: Khelben Blackstaff @ 2010-04-22 16:13 UTC (permalink / raw)
To: linux-raid
> I believe that these disks only come in the "green" variety. I recently
> picked up a 1.5 tb version for testing and cheap bulk storage, and I
> would not suggest using them in a raid array because the green drives
> firmware automatically parks the head after 8 seconds of inactivity and
> reduces the rpm of the disk. The constant parking can quickly wear out
> the head under high use and there is no way to disable this "feature".
As previously mentioned wdidle utility can disable the head unloading.
> The specifications say it's good for 300,000 cycles, so do the
> math... getting 5 unloads per minute would lead to probable failure
> after 41 days. Granted that is about worst case, but still something to
> watch out for. In order to make it the entire 3 year warranty period,
> you need to stay under 11.4 unloads per hour. If you have very little
> IO activity, or VERY MUCH, then this is entirely possible, but more
> moderate loads in the middle have been observed to cause hundreds of
> unloads per hour.
WD mentions in the customer help (Answer ID 5357) that these newer drives
were validated to 1M load/unload cycles and not 300K.
>
> If you want to fix it, then wdidle3.exe worked for me. Search for:
>
> wdidle3_1_00.zip
>
It worked fine for me too. I had a Load Cycle Count of 59, 15 mins
after connecting the drive for the first time. After running wdidle
the Count stopped at 104 and never increased again. I now have 11 hours
With wdidle 1.00 disabling the timer did nothing. I had to set the timer to
a large value. There is a newer version though (1.03) that supports
these new drives better.
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-22 16:13 ` Use of WD20EARS with MDADM Khelben Blackstaff
@ 2010-04-22 18:16 ` Simon Matthews
2010-04-22 19:44 ` Phillip Susi
1 sibling, 0 replies; 52+ messages in thread
From: Simon Matthews @ 2010-04-22 18:16 UTC (permalink / raw)
To: Khelben Blackstaff; +Cc: linux-raid
On Thu, Apr 22, 2010 at 9:13 AM, Khelben Blackstaff
<eye.of.the.8eholder@gmail.com> wrote:
>> I believe that these disks only come in the "green" variety. I recently
>> picked up a 1.5 tb version for testing and cheap bulk storage, and I
>> would not suggest using them in a raid array because the green drives
>> firmware automatically parks the head after 8 seconds of inactivity and
>> reduces the rpm of the disk. The constant parking can quickly wear out
>> the head under high use and there is no way to disable this "feature".
>
> As previously mentioned wdidle utility can disable the head unloading.
>
>> The specifications say it's good for 300,000 cycles, so do the
>> math... getting 5 unloads per minute would lead to probable failure
>> after 41 days. Granted that is about worst case, but still something to
>> watch out for. In order to make it the entire 3 year warranty period,
>> you need to stay under 11.4 unloads per hour. If you have very little
>> IO activity, or VERY MUCH, then this is entirely possible, but more
>> moderate loads in the middle have been observed to cause hundreds of
>> unloads per hour.
>
> WD mentions in the customer help (Answer ID 5357) that these newer drives
> were validated to 1M load/unload cycles and not 300K.
>
>>
>> If you want to fix it, then wdidle3.exe worked for me. Search for:
>>
>> wdidle3_1_00.zip
>>
I just looked at a WD green drive that was in a RAID1 set for several
months. The drive is an older model -- WD10EADS, but I think similar.
I did not use the widdle utility and the S.M.A.R.T. data reports a
load cycle count of 42. The RAID set held user directories, so it was
either under constant access (daytime) or not at all (nighttime, apart
from during backups).
Simon
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-22 16:13 ` Use of WD20EARS with MDADM Khelben Blackstaff
2010-04-22 18:16 ` Simon Matthews
@ 2010-04-22 19:44 ` Phillip Susi
2010-04-22 23:23 ` Mark Knecht
1 sibling, 1 reply; 52+ messages in thread
From: Phillip Susi @ 2010-04-22 19:44 UTC (permalink / raw)
To: Khelben Blackstaff; +Cc: linux-raid
On 4/22/2010 12:13 PM, Khelben Blackstaff wrote:
> WD mentions in the customer help (Answer ID 5357) that these newer drives
> were validated to 1M load/unload cycles and not 300K.
Weird, I wonder if they just forgot to update the spec sheet for the
drive then since it says 300k.
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-22 19:44 ` Phillip Susi
@ 2010-04-22 23:23 ` Mark Knecht
2010-04-23 0:03 ` Richard Scobie
2010-04-23 3:44 ` Phillip Susi
0 siblings, 2 replies; 52+ messages in thread
From: Mark Knecht @ 2010-04-22 23:23 UTC (permalink / raw)
To: Phillip Susi; +Cc: Khelben Blackstaff, linux-raid
On Thu, Apr 22, 2010 at 12:44 PM, Phillip Susi <psusi@cfl.rr.com> wrote:
> On 4/22/2010 12:13 PM, Khelben Blackstaff wrote:
>> WD mentions in the customer help (Answer ID 5357) that these newer drives
>> were validated to 1M load/unload cycles and not 300K.
>
> Weird, I wonder if they just forgot to update the spec sheet for the
> drive then since it says 300k.
Validated doesn't mean spec'ed, right? So they spec it at 300K but
they test and find that some units go to 1M, so they report that some
units go to 1M but they keep the spec at 300K. We feel more
comfortable with the operation but if the drive fails at 400K then
they are within their rights to say the drive lived a useful life and
replace or not replace it as they see fit.
I have 7 of these WD10EARS green drives, one in the machine I reported
on that's 350 miles away, but 6 sitting in boxes here. I think I'll
look into using the WD program to modify one or two of them and then
start some testing.
Keep in mind that just because we possibly solve the load cycle count
problem doesn't mean that the drive will work for RAID. WD has also
stated that these drives don't have any TLER features.
I'll report back how that goes, but I don't know exactly when.
Cheers,
Mark
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-22 23:23 ` Mark Knecht
@ 2010-04-23 0:03 ` Richard Scobie
2010-04-23 1:29 ` Mark Knecht
2010-04-23 3:49 ` Phillip Susi
2010-04-23 3:44 ` Phillip Susi
1 sibling, 2 replies; 52+ messages in thread
From: Richard Scobie @ 2010-04-23 0:03 UTC (permalink / raw)
To: Mark Knecht; +Cc: Phillip Susi, Khelben Blackstaff, linux-raid
Mark Knecht wrote:
> Keep in mind that just because we possibly solve the load cycle count
> problem doesn't mean that the drive will work for RAID. WD has also
> stated that these drives don't have any TLER features.
I completely agree and wonder if some have unrealistic expectations.
The WD20EARS is listed as a desktop drive and is cheap - $/GB.
The RE-GP, (if you need the green features, otherwise RE series),
version WD2002FYPS cost more and has the following extra features over
the WD20EARS:
RAFF - vibration handling
TLER - RAID optimised error handling
An MTBF spec and the comment
"Each drive is put through extended burn-in testing with thermal cycling
to ensure reliable operation."
Which presumably helps weed out infant mortality cases. It is also
interesting to note that while all the RE series drives have a spindle
support bearing at both ends, only the 2TB desktop drive does.
In light of all the above, you get what you pay for and need to adjust
expectations accordingly.
Regards,
Richard
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-23 0:03 ` Richard Scobie
@ 2010-04-23 1:29 ` Mark Knecht
2010-04-23 3:49 ` Phillip Susi
1 sibling, 0 replies; 52+ messages in thread
From: Mark Knecht @ 2010-04-23 1:29 UTC (permalink / raw)
To: Richard Scobie; +Cc: Phillip Susi, Khelben Blackstaff, linux-raid
On Thu, Apr 22, 2010 at 5:03 PM, Richard Scobie <richard@sauce.co.nz> wrote:
> Mark Knecht wrote:
>
>> Keep in mind that just because we possibly solve the load cycle count
>> problem doesn't mean that the drive will work for RAID. WD has also
>> stated that these drives don't have any TLER features.
>
> I completely agree and wonder if some have unrealistic expectations.
>
> The WD20EARS is listed as a desktop drive and is cheap - $/GB.
>
> The RE-GP, (if you need the green features, otherwise RE series), version
> WD2002FYPS cost more and has the following extra features over the WD20EARS:
>
> RAFF - vibration handling
> TLER - RAID optimised error handling
> An MTBF spec and the comment
>
> "Each drive is put through extended burn-in testing with thermal cycling to
> ensure reliable operation."
>
> Which presumably helps weed out infant mortality cases. It is also
> interesting to note that while all the RE series drives have a spindle
> support bearing at both ends, only the 2TB desktop drive does.
>
> In light of all the above, you get what you pay for and need to adjust
> expectations accordingly.
>
> Regards,
>
> Richard
<hehe!>
If drives always worked there wouldn't be much reason for RAID1, would
there? ;-)
Cheers,
Mark
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-23 0:03 ` Richard Scobie
2010-04-23 1:29 ` Mark Knecht
@ 2010-04-23 3:49 ` Phillip Susi
1 sibling, 0 replies; 52+ messages in thread
From: Phillip Susi @ 2010-04-23 3:49 UTC (permalink / raw)
To: Richard Scobie; +Cc: Mark Knecht, Khelben Blackstaff, linux-raid
On Fri, 2010-04-23 at 12:03 +1200, Richard Scobie wrote:
> "Each drive is put through extended burn-in testing with thermal cycling
> to ensure reliable operation."
>
> Which presumably helps weed out infant mortality cases. It is also
> interesting to note that while all the RE series drives have a spindle
> support bearing at both ends, only the 2TB desktop drive does.
Indeed. Pretty standard manufacturing process to weed out the infant
mortality. Interesting to find out that they don't bother with these
drives. That might explain why my first one died after ~24 hours of
operation. What still pisses me off is that it died so hard my bios
wouldn't even post with the thing plugged in, let alone let me read any
SMART data from it.
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-22 23:23 ` Mark Knecht
2010-04-23 0:03 ` Richard Scobie
@ 2010-04-23 3:44 ` Phillip Susi
1 sibling, 0 replies; 52+ messages in thread
From: Phillip Susi @ 2010-04-23 3:44 UTC (permalink / raw)
To: Mark Knecht; +Cc: Khelben Blackstaff, linux-raid
On Thu, 2010-04-22 at 16:23 -0700, Mark Knecht wrote:
> Validated doesn't mean spec'ed, right? So they spec it at 300K but
> they test and find that some units go to 1M, so they report that some
> units go to 1M but they keep the spec at 300K. We feel more
> comfortable with the operation but if the drive fails at 400K then
> they are within their rights to say the drive lived a useful life and
> replace or not replace it as they see fit.
Specifications are based on validation; they don't have anything to do
with the warranty. If one drive made it to a million, who cares? The
question is will the majority of them? Just because one drive made it
to a million hours before failing does not mean a million hours is the
MTBF, nor does a drive failing at half MTBF have anything to do with
whether it is still under warranty or not. The warranty is expressed in
real time duration, not hours of operation or head unload count.
The question is, if you have a dozen drives in a raid array, is there a
good chance that none will fail after they hit 300,000 unloads. If the
answer to that is no, and it looks like they will reach 300,000 unloads
in a few months, then you probably don't want to use those drives.
> Keep in mind that just because we possibly solve the load cycle count
> problem doesn't mean that the drive will work for RAID. WD has also
> stated that these drives don't have any TLER features.
TLER sounds like marketing department BS to me. ANY drive should be
able to retry a read plenty of times and give up in less than 7 seconds
without some special feature. I've never used a drive with this
"feature" and never had mdadm kick a drive offline just because it
developed a bad sector. Usually I notice from a smart check and force a
write to the sector and the drive remaps it.
^ permalink raw reply [flat|nested] 52+ messages in thread
* Re: Use of WD20EARS with MDADM
2010-04-21 13:20 ` Bill Davidsen
` (2 preceding siblings ...)
2010-04-21 15:14 ` Phillip Susi
@ 2010-04-21 15:52 ` Simon Matthews
2010-04-21 19:24 ` Richard Scobie
4 siblings, 0 replies; 52+ messages in thread
From: Simon Matthews @ 2010-04-21 15:52 UTC (permalink / raw)
To: Bill Davidsen; +Cc: Phillip Susi, st0ff, linux-raid
On Wed, Apr 21, 2010 at 6:20 AM, Bill Davidsen <davidsen@tmr.com> wrote:
> Phillip Susi wrote:
>>
>> On 4/14/2010 3:53 PM, Bill Davidsen wrote:
>>
>>>
>>> Stefan *St0fF* Huebner wrote:
>>>
>>>>
>>>> Am 25.03.2010 18:45, schrieb Asdo:
>>>>
>>>>>
>>>>> David Lethe wrote:
>>>>>
>>>>>>
>>>>>> On Mar 25, 2010, at 11:20 AM, Andrew Dunn wrote:
>>>>>>
>>>>>>>
>>>>>>> Recently Dell is selling their WD20EARS (2TB) for 90$
>>>>>>>
>>>>>>> This being one of the best PPG (Price Per Gigabyte) snatches I have
>>>>>>> seen, I was considering buying 8 of them for an mdadm array.
>>>>>>>
>>>>>>> Has anyone had experience with these drives?
>>>>>>>
>>
>> I believe that these disks only come in the "green" variety. I recently
>> picked up a 1.5 tb version for testing and cheap bulk storage, and I
>> would not suggest using them in a raid array because the green drives
>> firmware automatically parks the head after 8 seconds of inactivity and
>> reduces the rpm of the disk. The constant parking can quickly wear out
>> the head under high use and there is no way to disable this "feature".
>>
>
> I hear this said, but I don't have any data to back it up. Drive vendors
> aren't stupid, so if the parking feature is likely to cause premature
> failures under warranty, I would expect that the feature would not be there,
> or that the drive would be made more robust. Maybe I have too much faith in
> greed as a design goal, but I have to wonder if load cycles are as
> destructive as seems to be the assumption.
>
> I'd love to find some real data, anecdotal stories about older drives are
> not overly helpful. Clearly there is a trade-off between energy saving,
> response, and durability, I just don't have any data from a large population
> of new (green) drives.
It's not hard data, but there was discussion of a similar issue fours
years ago related to load/unload cycles on laptops under Ubuntu:
https://bugs.launchpad.net/ubuntu/+source/acpi-support/+bug/59695
Simon
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 52+ messages in thread* Re: Use of WD20EARS with MDADM
2010-04-21 13:20 ` Bill Davidsen
` (3 preceding siblings ...)
2010-04-21 15:52 ` Simon Matthews
@ 2010-04-21 19:24 ` Richard Scobie
4 siblings, 0 replies; 52+ messages in thread
From: Richard Scobie @ 2010-04-21 19:24 UTC (permalink / raw)
To: Bill Davidsen; +Cc: Phillip Susi, st0ff, linux-raid
Bill Davidsen wrote:
> I'd love to find some real data, anecdotal stories about older drives
> are not overly helpful. Clearly there is a trade-off between energy
> saving, response, and durability, I just don't have any data from a
> large population of new (green) drives.
I can add some very limited, (6 drives) observations on the 2GB RE
version - WD2002FYPS.
These are green drives and have been running continuously in an md RAID5
for eight months without any problems and no drive settings have been
altered. This array is idle except for a nightly rsync session for an
hour or so.
Looking at the SMART data, load cycle count, is 23862 which sounds like
a lot, but the current VALUE entry is 193 (and started at 200 8 months
ago) and the THRESHOLD value is 0.
So, if I am interpreting this trend correctly, the THRESHOLD value will
be reached in about 682,000 load cycles and about 228 months.
Regards,
Richard
^ permalink raw reply [flat|nested] 52+ messages in thread