ST32000644NS - Nightmare

[pinoy]

Just relaying on experience on these.
Its crazy receiving 2 x enterprise hdd both failing within just over a year on a RAID 1+0 set up! Client thinks both failed almost simultaneously (hhmmm interesting. I am inclined not to believe this).

Clients MISTAKENLY THINK it is safe to have a Raid 1 or 5 or 6..etc and disregard backups!!!! Good Verified backups are a MUST!!!!!

Going to be an expensive recovery for parts. I cannot rely on clients words when they say both failed at same time (only i can think of is if somebody has sabotaged the server and kicked it hard enough while its running).

Did the usual diagnostics but, both required opening in a cleanroom. One has a head crash and has scraped the platter surface. Other drive probably has bad heads, clicking and powers off..

See how it goes when donor arrives tomorrow.

quite challenging drives to work on these Contellation ES.

How's your Seagate ES experiences??

[Nick_CT]

I often wonder what some of these people do. Last week had a Proliant rack mount with 3x Seagate SAS 10k drives, 2 of the 3 had head crashes and massive platter damage. User said they were 'working fine' the day before. So they both failed in a physical way all by themselves, at the same time? Yeah right.

[BlackST]

Unless there was intentional or accidental damage, people usually repeat the 'simultaneous fail' bullshit (donkeyshit...). In reality : servers run until smoke comes out or they hear a grinding sound out of the rack because MANAGEMENT, PREVENTIVE MAINTENANCE and SPARE PARTS (oh yes, there was a lit LED ... 'alarm' ... Forgot it) do COST. If only the server was the boss' car...

[warnerr]

Prolients are really HEAVY- best guess is server dropped while servicing and no one will admit it. Pure guesswork.

[pinoy]

...servers run until smoke comes out or they hear a grinding sound out of the rack because MANAGEMENT, PREVENTIVE MAINTENANCE and SPARE PARTS (oh yes, there was a lit LED ... 'alarm' ... Forgot it)...

right on the money... raid controller indicated one of the drive have failed a while back.. Nothing was done till the other failed. They are off a Proliant ML 350 G6. After further speaking to the clients on call "techie" or should I say will be EX-TECHIE.. Server was locked up in a secured comms room so it's out of the way from being knocked around. He continues, "apparently there was a backup being done on D: drive" !!!!!!

Well i tell him whoever is responsible for this Server is TOTALLY SCREWED!

I see a possible lawsuit coming on! D: drive is part of the same RAID1+0 array!!!!!

The pressure is on!
Drive 1 with the huge deep scrapes on the platter is a no go! So much metallic debris I removed. The grooves will definitely kill heads on power up. (Where is that magic salvation nano putty when u need one ) jok.

Drive 2 is the only hope! Hoping all surface is ok!!!! But with previous experience, these 8 head 4 platter drives usually die dramatically, usually with head crashes).

These could possibly be from a bad batch, or perhaps with the extensive usage on them coupled with dust/material accumulation on the vents = overheated environment... good way to kill drives!

better get some sleep now. nitez!

[BlackST]

Sorry to hear this, as I stated I see these scenarios almost every week.

Unfortunately I can confirm that enterprise drives 7-8 out of 10 come to DR into dramatical condition , where a "CORONER" rather than a "DR ENGINEER" is needed

It is pointless to look for the cause of "death" but in my stats it is simple end of life , then less often overheat or possible contamination.

Good luck for your recovery, hope at least some relevant data will come out from the working drive - don't have any suggestion about the other drive(s) because if media is in such mess it is "donor suicide" even to think trying a headswap or so.

[pinoy]

... Cut to the chase...

After a couple of headstack replacement, some acelab tool & lots of terminal scripting magic.... Drive is functional and FINALLY data is accessible and now Imaging, very slowly (3% in 2hrs)!!!!!!! May require more tweak settings from the imager to handle the occasional bunches of contiguous bad blocks..

Also had to put it in an enclosed fan cooled rack as the drive temp was climbing quite uncomfortably.

This priority case recovery has cost me a few thousand grey hairs, eye twitching strain and a possible heart murmur from drinking a couple of litres of redbull!!!!

Oh yeah, these drives were manuf 2011, with 5yrs wnty. hehe!