ack! i have some wd aaks hard drives!

[tmp31416]

(i hope this is the right forum for this)

 

...so i stumbled on the "does not work!" list, which includes the wd aaks sata drives.

i have three of them in my server, already configured to be part of my main share.

oops.

i knew i wanted to eventually replace all 500gb drives with 1tb ones, but i might have to hasten the pace on this.

so the silly question "du jour" is: can i powerdown my server (i have pimped its two incarnations in the lounge already (ahem, can i say this?)), replace one 500gb (wd aaks) drive with a 1tb one (wd ears) in the same slot of its drive, turn the server back on and have unraid automagically recreate the drive?

that sounds too good to be true, better to ask and make a fool of myself rather than lose my data...

 

cheers.

 

[Joe L.]

(i hope this is the right forum for this)

 

...so i stumbled on the "does not work!" list, which includes the wd aaks sata drives.

i have three of them in my server, already configured to be part of my main share.

oops.

i knew i wanted to eventually replace all 500gb drives with 1tb ones, but i might have to hasten the pace on this.

so the silly question "du jour" is: can i powerdown my server (i have pimped its two incarnations in the lounge already (ahem, can i say this?)), replace one 500gb (wd aaks) drive with a 1tb one (wd ears) in the same slot of its drive, turn the server back on and have unraid automagically recreate the drive?

that sounds too good to be true, better to ask and make a fool of myself rather than lose my data...

 

cheers.

 

As long as your parity drive is a 1Gig drive you can do exactly as you said.  Just be certain to use the "Start" button to begin the re-construction process.  Do NOT use the button labeled as "restore" as it has nothing to do with restoring data. It is badly labeled.  It is actually a "Delete Disk Configuration and Parity" button. 

 

If you current parity disk is not already 1Gig or larger, then first power down replace the parity disk with a 1Gig, power up, press "Start" and let it build parity on the 1Gig drive.  Then you can do as you said, power down, replace a data drive, power up, press "Start" let it re-construct the old contents of the replaced drive onto the new.  Then repeat as needed for each disk in turn.

 

During the entire process you should ONLY need press the "Start" button.  You do not need to format any disks.  If a disk shows as un-formatted DO NOT PRESS THE FORMAT BUTTON.  Stop the array and re-start it ad it should get the screen into sync.

 

Joe L.

[prostuff1]

If the drives are not giving you trouble do not replace them.  I am running 2 500GB drives in my server and of all the drives I have then are the most reliable.  They have not given me any trouble and are both "over" the threshold for hours in use.  I think they are reporting 30000 hours each

[lionelhutz]

I started my array with 3 of them and they worked fine for the time they were in there. I upgraded them because I wanted more space and it gave me some extra SATA drives to use in other projects. If I hadn't wanted to have some extra drives around for other uses they'd still be in the server.

 

So, if you're not having problems with the server then don't worry about them.

 

Peter

 

[tmp31416]

well, everything is running fine at the moment, but since the bandwidth of roaming hard drives is better than anything mere mortals can reasonably afford when it comes to transferring _massive_ amounts of data (*) between distant networks, i thought i could kill three birds with one stone.  

 

first, i could get this aaks drive into a small enclosure that i could lend to friends that have lots of data to store "off site";

secondly, since ears drives appear to be less problematic than the aaks, i should increase reliability -- or lessen the chance of "misfire" in my server;

lastly, i'd gain 500gb of space in my main share each time i swap a drive.

 

so, as Joe L. said, when i swap the 500gb aaks for a 1.0tb ears and i turn on the server, the array will not be started and "disk x" will be in a "funny" state, right?  then, just by clicking on the "start" button (to start the array), the drive that will become the new "disk x" will be automatically cleared, formatted and populated with the data re-constructed from the parity & other drives, yes?

 

again, i prefer to ask the obvious and look like a fool instead of staying quiet in a futile attempt to look intelligent, making some bad assumptions, only to lose a tonne of data...

 

cheers.

 

(*) a variation on "never underestimate the bandwidth of a station wagon loaded with tapes".

 

[lionelhutz]

Yup, that's basically what you do. I think there is a check box associated with the start button that says something like "are you sure" that you have to first click before you can use the start button.

 

Peter

 

[jupilerman]

...so i stumbled on the "does not work!" list, which includes the wd aaks sata drives.

So what ?

I have 6 of them inside my unRAID for about 2 1/5 years now and I never had any trouble.

parity 	WDC_WD10EACS-00D_WD-WCAU46319917 	* 	976,762,552 	-             	0 	0 	0
disk1 	WDC_WD5000AAKS-0_WD-WMASY2679981 	* 	488,386,552 	10,500,500 	0 	0 	0
disk2 	WDC_WD5000AAKS-00A7B0_WD-WMASY2021827 	* 	488,386,552 	10,460,800 	0 	0 	0
disk3 	WDC_WD10EADS-00L_WD-WCAU4A408744 	* 	976,762,552 	10,087,996 	0 	0 	0
disk4 	WDC_WD5000AAKS-0_WD-WMASY3297321 	* 	488,386,552 	10,162,984 	0 	0 	0
disk5 	WDC_WD5000AAKS-0_WD-WCAPW5754626 	* 	488,386,552 	10,995,312 	0 	0 	0
disk6 	WDC_WD10EADS-00L5B1_WD-WCAU4A425557 	* 	976,762,552 	9,906,436 	0 	0 	0
disk7 	WDC_WD5000AAKS-0_WD-WMASY3311462 	* 	488,386,552 	14,953,016 	0 	0 	0
disk8 	WDC_WD7500AAKS-0_WD-WCAPT0051250 	* 	732,574,552 	41,316,636 	0 	0 	0

[neilt0]

...so i stumbled on the "does not work!" list, which includes the wd aaks sata drives.

So what ?

I have 6 of them inside my unRAID for about 2 1/5 years now and I never had any trouble.

 

Yeah, I was going to say the same thing. I have maybe 10 in my two servers, and no problems so far, over a year later.

[MortenSchmidt]

...so i stumbled on the "does not work!" list, which includes the wd aaks sata drives.

So what ?

I have 6 of them inside my unRAID for about 2 1/5 years now and I never had any trouble.

 

Yeah, I was going to say the same thing. I have maybe 10 in my two servers, and no problems so far, over a year later.

 

Enjoy it while they last. I own 3 and they have failed 4 times. First, they all failed within two months, the youngest one was 2.5 years the oldest was just one month shy of the 3 year warranty. Since they came back, one has failed again and I am now only using the two remaining drives for off-site backup storage. Will prioritize replacing them soon even for that purpose.

[neilt0]

...so i stumbled on the "does not work!" list, which includes the wd aaks sata drives.

So what ?

I have 6 of them inside my unRAID for about 2 1/5 years now and I never had any trouble.

 

Yeah, I was going to say the same thing. I have maybe 10 in my two servers, and no problems so far, over a year later.

 

Enjoy it while they last. I own 3 and they have failed 4 times. First, they all failed within two months, the youngest one was 2.5 years the oldest was just one month shy of the 3 year warranty. Since they came back, one has failed again and I am now only using the two remaining drives for off-site backup storage. Will prioritize replacing them soon even for that purpose.

 

The original post was about compatibility, not reliability. Your post is claiming a reliability issue, but you only cite your own experiences with a tiny sample of drives. This is not statistically relevant, and I can't find a major problem with these drives after some searching.

 

I have had no WD drives fail, yet I've had 3 Seagate drives fail, 2 within one week. Does that mean all Seagate drives have design faults? No. Because that's too small a sample.

 

Below, I've pasted the SMART summaries for my two servers. I know one of the WD drives is failing(ish), but the reallocated sector count hasn't increased in years, so I'll only replace it if really starts dying. It's also had the most use (not just in hours, but in getting hammered before going in to the unRAID array).

 

If any experts want to comment on the SMART summaries, please go ahead.

 

 

[tmp31416]

thanks to Joe L. & Lionel Hutz (who is a better sysadmin than lawyer -- you might want to quit that day job!) answering my question on how to do it, i swapped one of my 500 gb aaks for a 1tb ears and had unraid do its magic.  despite working in data centres that would make some around here have to change their underwear, i did find it exceedingly cool -- hey, it's not just rebuilding a "missing" drive, it's upgrading it at the same time!

 

just for that made me happy to have been contaminated with the "unraid bug".    :-D

 

thanks again, chaps!

 

[MortenSchmidt]

To add to what's already being said - You should also do a No-correct parity check after upgrading that disk. See my thread in the support forum if You want to know why.

http://lime-technology.com/forum/index.php?topic=12884.0

 

In short, I was as excited as You are, after having performed my first disk upgrade. That is untill happened to run a parity check a few days later and found repeatable/persistent parity errors (thankfully I performed a no-correct parity check and not a regular sync parity check). I'd appreciate the attention from the senior forum members who have commented on this thread over in my own thread. Thanks!