Losing disks and my sanity

[CharleyVarrick 283]

Test #1, a brand new, out of the bag 8tb Seagate drive (the previous lasted an hour), same sata port

In disk mgmt, t is recognized and request initialization (GPT)

I will instruct DrivePool that its part of the movie pool

 

for clarity, I will refer to this latest disk as SG8b

 

I will install second to last disk (SG8a) in my secondary server momentarly

Edited February 3, 2018 by jlr19

[CharleyVarrick 283]

I'll be damned! SG8a (second to last disk, that worked about an hour) is a no-show in backup server.

 

Meanwhile SG8b is purring along in main server, I will throw a few hundred gigabites its way.

 

If it just a DOA, its odd how it mimick exactly the behavior of the 1 year old WD blue it replaced.

[PenkethBoy 2068]

have you tested with the wd diagnostic tool to see if it can see the disc(s) and if so run the tests

 

It may well be coincidence with SG8a

 

another more worrying alternative is there might be a power regulation issue thats frying the disks - unlikely though

[PenkethBoy 2068]

did you return the wd disks under warranty? - 2 years on a wd blue 

[CharleyVarrick 283]

have you tested with the wd diagnostic tool to see if it can see the disc(s) and if so run the tests

 

It may well be coincidence with SG8a

 

another more worrying alternative is there might be a power regulation issue thats frying the disks - unlikely though

Data Lifeguard does not detect it.

I put my hand on it and its definitely "trying", its hot (but not burning hot) and turning.

 

Considering the number of drives I went through in the last 18 months (12 ! ! !), power issue is a worrying prospect indeed

Edited February 3, 2018 by jlr19

[Senna 368]

Alarm bells should have been ringing much earlier, if same system already has nuked 12 drives

Or are those from different systems ?

[CharleyVarrick 283]

I am not kidding you, my latest drive just vanished in thin air within merely 30 minutes

I am shocked!

[CharleyVarrick 283]

Alarm bells should have been ringing much earlier, if same system already has nuked 12 drives

Or are those from different systems ?

2 similar systems

Main server: 8

Backup server: 4

Work pc, same 2 drives in for 7 years and counting (all connected on same electrical circuit)

 

Servers have each 6 sata ports, with 4 port expansion card added (to each) in the last 18 months, as well as both running StableBit DrivePool/Scanner combo

BTW, StableBit Scanner is useless in this scenario as 1 minute the drive is healthy, the next minute its gone

[CharleyVarrick 283]

I rebooted, the last drive came back (as expected) but went south again within 10 minutes.

[CharleyVarrick 283]

I am bringing this PoS to the computer shop right now (and its not close by).

I'll report back later

[Senna 368]

Seems that your 12V lines of your power supply aren't correctly regulated, causing ripples in current, delivered to your drives.

 

I should definitely examine the power supply or just replace it with a quality one, which is proven to handle a clean and steady current to your hardware and has some reserve.

[PenkethBoy 2068]

I would replace the power supply, m/b (which will mean a new chip) and probably memory as well - depending what you pick etc

 

but if it is killing drives this quick then... not much choice

 

scanner is not design to inform you of this sort of failure - as once the drive disappears it thinks you removed it

[Swynol 375]

i had this exact same issue with hard drives just dropping off randomly from drive pool. 

 

it was completely random what disk it was so i knew the disks were fine. turned out to be the power supply. I was running a 600w PSU with about 10 drives.

 

i added a second power supply using a paper clip on the 24pin to turn the psu on and just run 4-8 drives off the PSU. the origanal PSU then ran everything else until i upgraded my PSU to a 850w gold.

[Senna 368]

Indeed, re-occuring HDD failures are mostly related to a cheap ass PSU, overloaded PSU or a failing PSU.

 

Most people underestimate the importance of a good quality PSU, that is also suited for the (future) load its gonna get, in systems that run 24hrs/day.

Edited February 3, 2018 by Senna

[CharleyVarrick 283]

Back from the computer store, just finished replacing psu with a new Corsair CX750m (twice the price of the original one, but cheap compared to a pile of disk drives), now the moment of truth...

 

EDIT: Powered up, it didn't blew up, all drives are accounted for, at least for now.

I'll leave it running and keep a close eye on it.

 

I forgot to mention a detail that points to the psu as the culprit; once in a while, I was catching my monitor's brightness flickering.

 

Please-please-please, stop eating my drives, you silly computer.

Edited February 3, 2018 by jlr19

[Guest asrequested]

[CharleyVarrick 283]

That sums it up!

Hi Doofus, long time no chat (my fault)

[Guest asrequested]

I trust that your problem is solved?

[Senna 368]

Are you using a surge protector for your computer equipment ?

Because your electrical house wiring could also be a factor, for the problems you had, resulting in a failing PSU.

[Happy2Play 9780]

I have a machine that does the same thing, it runs 16 drives fine but add drive 17 and randoms drives just fall offline.  Always suspected PSU but never got around to upgrading it yet as I just keep replacing smaller drives with bigger ones.

[CharleyVarrick 283]

I trust that your problem is solved?

1 hour in, so far so good, and the new psu runs much quieter than the old one (thought the racket was from chassis fan).

[CharleyVarrick 283]

I have a machine that does the same thing, it runs 16 drives fine but add drive 17 and randoms drives just fall offline.  Always suspected PSU but never got around to upgrading it yet as I just keep replacing smaller drives with bigger ones.

Hi Happy.

I have a drawer full of smaller and not so smaller drives that I (fortunately) kept. Holiday project to test carefully them. Afew might actually be toast, but I'll bet I'll recover 10-20 tb's in the process.

[Guest asrequested]

I always use EVGA PSUs

 

https://www.amazon.com/EVGA-SuperNOVA-Modular-Warranty-220-G2-0850-XR/dp/B00IKDETOC/ref=sr_1_4?ie=UTF8&qid=1517688726&sr=8-4&keywords=evga+g2+power+supply

[CharleyVarrick 283]

Are you using a surge protector for your computer equipment ?

Because your electrical house wiring could also be a factor, for the problems you had, resulting in a failing PSU.

For sure, and a good one too! My work pc seems to agree with its 2 drives well into their 6th or 7th years of work without a fuss.

This is the second time a ThermalTake product fails me miserably, and probably the last time too!

[CharleyVarrick 283]

I always use EVGA PSUs

 

https://www.amazon.com/EVGA-SuperNOVA-Modular-Warranty-220-G2-0850-XR/dp/B00IKDETOC/ref=sr_1_4?ie=UTF8&qid=1517688726&sr=8-4&keywords=evga+g2+power+supply

Ouch, those are pricey!

$220cdn, the one I bought earlier today was $120, as for the one I replaced must have been around $75