Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[sevenlayermuddle]

Thanks, but this is a purely temporary issue, providing it lasts another day or two I've no longer any plans to repair it.

When the migration to new box is complete, the next port of call for the old one will be my local authority recycling facility.  I may rescue a few bits and bobs like the CPU and memory,  to gather dust in a cupboard in case they ever prove useful  but the PSU and mobo are now doomed.  As is the affected disk, once it is fully erased.

[burakkucat]

As is the affected disk, once it is fully erased.

And then gently caressed with a 14 lb hammer! 

[sevenlayermuddle]

As is the affected disk, once it is fully erased.

And then gently caressed with a 14 lb hammer! 

dd of=/dev/sda bs=1024k if=14lb ?

[burakkucat]

Yes, that would do it or even usage the shread command. [1]


[1] http://linux.die.net/man/1/shred

[NewtronStar]

remove the top cover 4 screws access the platters and pull hard then enter the platters into openfire place/wood burner it's the only way to be sure 

[burakkucat]

I believe that some disk manufacturers actually make the platters out of glass. 

[NewtronStar]

I believe that some disk manufacturers actually make the platters out of glass. 

Ah well you can always use the Microwave oven


https://www.youtube.com/watch?v=xe8fOCWL2sU

[loonylion]

remove the top cover 4 screws access the platters and pull hard then enter the platters into openfire place/wood burner it's the only way to be sure 

my dad drills through the platters

[sevenlayermuddle]

Getting back to the original problem (burn up) I just noticed

HP have know about the problem since 2012 and that it exists on my specific server, an ML110 G5

http://h30499.www3.hp.com/t5/ProLiant-Servers-ML-DL-SL/ML110-G5-servers-catch-on-fire/m-p/5807531#M132300
 

The link to a customer advisory, at the end of that thread, seems broken so no idea what it says.   Considering the number of these machines deployed world wide, I wonder if any actual full-blown building fires have resulted?   

[guest]

I'd get onto HP UK about it were I you as they might refund the price (or extend/enhance the warranty) of your new machine - and with the cashback you'd be up on the deal .

Worth a try I'd say - personally I'd take the approach that you probably wouldn't have bought HP again if you'd known about it & its going to take something for them to earn your trust again.

Oh & the customer advisory notice does exist but after HP's recent split it exists on a different server :

http://h20564.www2.hpe.com/hpsc/doc/public/display?docId=emr_na-c03199591&lang=en-us&cc=us

A brief quote from it :

In very rare instances, the SATA Power Y-cable (HP Part Number 456853-001) which connects to location P6 on the power supply of a ProLiant ML110 G5 server and is used for powering a third and fourth SATA device, may fail even if it is not plugged into a SATA device. If this occurs, the server may not boot and the cable may become damaged inside the server chassis. In addition, the failure of the cable could, in rare instances, cause the attached peripheral device (SATA CD-ROM/HDD) to fail, fault, or become damaged during normal operation.

Sounds like that's the one, doesn't it?

Edit2 - "Cables that were built after May 2011 are properly functioning cables."

Get an email over to HP and ask for compo is my advice.

[sevenlayermuddle]

Oddly mine doesn't have that second sticker with a manufacturing date but I bought it in 2008 so clearly it predates the 'good' cables.

I may try and get in touch with them, you never know your luck.   I'm more annoyed that, despite the fact I went to all the trouble of registering it with them, they never bothered to notify me of such a safety hazard.  I guess it's possible they did send me an email  and it went to spam, my hunch is they didn't.

[guest]

My reading of the customer advisory is that it only went out in the US as its a US phone number you call for the replacement part.

Its a clear manufacturing defect which has resulted in a fire for some customers so HP UK will have to come up with compo or run the risk you take them to small claims court (where you would win, been there & got that T-shirt myself with a Shuttle box some years ago).

Get an email over to them - you should push for at least 3 years next business day onsite warranty IMHO (try for 5 years). I can't see them turning that down as its effectively a zero-cost option for them if they're confident in their hardware.

[sevenlayermuddle]

There's been a sting in the tail of this episode.

I've pretty much finished with the old server, data recovery/migration all completed and successful.   I was a bit worried about whether the disk attached to the burning cable might be damaged but, while it is destined recycling owing to dodgy looking PCB tracks, it has behaved impeccably.

But another disk from the old system, a 2TB Hitachi Deskstar, I noticed had picked up 2 bad sectors by the time I installed it in the new rig.   In the space of two weeks, it has grown to nearly 300 bad sectors.   

Not a huge problem as nothing on that disk mattered too much.   But co-incidence?   Probably not.   The old server had disks laid horizontally, one above the other.    This one was directly above the burning Sata and now I look closely, there are scorch marks on the casing.   

[guest]

/nag mode on

Did you email them?

This has cost you money & time and its a known defect.

/nag mode off

[sevenlayermuddle]

/nag mode on

Did you email them?

This has cost you money & time and its a known defect.

/nag mode off

Not yet.  Undecided TBH, and a tiny smidgen conflict of conscience... the system was 7 years old and I'd been planning to replace it soon anyway, with something more energy efficient.

And I've already has confirmation that the cash back's going through which reduces the financial hit.   If they did offer me any £££ compensation I'd have to be open and honest about the cash back.

You're right though.