Dead HDD?

I understand this means I should back up my shit and it's about to die.

But what exactly does this mean?

Other urls found in this thread:

backblaze.com/blog/hard-drive-failure-rates-q3-2017/
twitter.com/NSFWRedditGif

I'm also interested.

it means exactly as you said, your shit is about to die so back it up.

I'm no expert but looks like it has already passed the limit for sector reallocation and has no more spare sectors, if this happened it means the drive is pretty much so fucked it already wasted every extra sector, expect corrupt data.

What's weird is that the power on hours is only 4k hours and the power on count 159, that's pretty new for an HDD. I guess maybe toshiba is shit or 3tb drives are failure prone.

A single sector on the platter died,whichever file that had data written on that sector may be corrupted. Once a single one appears, in many cases bad sectors count would multiple rapidly in a very quick manner.

don't bother checking anything, copy paste the entire drive to a new one and check the files once they have been copied so you don't waste more time. You still can use it as a junk drive.

A strong neodymium magnet can probably restore the sectors.

Create a backup image of your hard drive, replace the hard drive, restore from backup image. Done

>HURR LE SEAGATE IS LE EPIK BAD XD
>toshiba lasts 4000 hours

Looks like Sup Forums is wrong yet again.

Ive had 4 of that exact model die on me

backup your shit now, last time i only managed to copy about 40% before the hdd wouldent even register anymore.

I'm not sure. A reallocated sector means your sector died and it moved the data to a spare sector. It seems you have one remaining spare sector.

What I'm curious about, is data corrupted when a sector gets reallocated and is there a way to check which data on Windows?

you need to elaborate more when you say things like this

Place magnet on face of large hammer. Apply hammer to damaged sectors.

Use ddrescue.
Now.

Data is already corrupt.
This will help you salvage as much as you can!

I have a 1tb drive from 2009, it has 58000 hours on it

It's still showing as healthy

I had 3x3tb drives die on me in a year, toshiba and seagate

The 1tb is WD

WD black is really the best.

>4374 hours
>=182 days
How? I have a drive nearing 30k hours and works fine.

your sample size is not convincing.

Toshiba is shit. All of the toshiba brand notebook we have at work had their OEM HDDs replaced

>OS drive on my pc got 5 reallocated sectors out of nowhere
>start worrying that it might fail at any minute
>5 years later it's still working and the 5 value hasn't gone up

feels good man

If you don't backup, it won't be feeling so good anymore.

Google published a study on their data center hard drives, the MTBF follows a "bathtub curve" - drives that failed usually failed within the first X months (the drive was shit) or after like 10 years (normal wear and tear)

...

you are like a little baby, watch this

I keep safe backups of everything on my PC nowadays
But I doubt it's going to fail now after all this time

>windows
$ sudo smartctl -a -d sat /dev/sdb | grep 'Hours'
9 Power_On_Hours 0x0032 062 062 000 Old_age Always - 38416
Imagine needing a GUI to read SMART

already beat you

...

Why do you think it's windows?

Ive had a bad sector for the last year. its backed up so no worries there. but ive just been letting it sit and do its thing. still 1 bad sector, hasnt gotten worse hasnt gotten better. idk man

im $ sudo smartctl -a -d sat /dev/sda | grep 'Hours'
9 Power_On_Hours 0x0032 044 044 000 Old_age Always - 41417
not gonna lie tho, i don't know how to use smarctl so I use gsmartcontrol instead but it doesn't show uptime

Because as far as I know that utility isn't developed for Linux.

You mean App?

As long as it doesn't go up you'll probably be fine
see

>turning off HDDs
try me

It really depends on how you use them. I have 12 of that exact model in my server without issues. Oldest one from what I remember was manufactured in 2013.

you know system reset counts to it
also
>barely 20k
lmao

install gentoo

Get on my level falamadingdongs

>low-end consumer trash
get the ultrastar

>Current pending
>956
Your shit is corrupt.

>chrome
>windows 10
>posts on a technology board

You are like a little babby to me

Watch this

Succ


Winfags wish they had data integrity this good.
All files green on checksums with both SSDs losing ~30k sectors.

>SEAGATE IS BAD GUISE
>HITACHI IS GOD TIER!

LMAO Sup Forums OFFICIALLY BLOWN THE FUCK OUT

I really should replace my torrent hdd.

Worse than that.
Your shit is already corrupt on that drive.

They may be torrents but if you're a lone seeder you'll be handing out 0.999
If you don't seed them then you'll have random fruit salad in your mkv files or they may fail to run at all.

>all these babbys

hnggggg

t. Arch user
Did not bother with too good drives for a toy-around server 2016 installation. Literally left this machine in the corner for a few years and viola.

Nah its fine.

You guys may want to give this a read,
backblaze.com/blog/hard-drive-failure-rates-q3-2017/

This shit is a brandless 7 years old hdd for my cfw ps3. I'm quite impressed it's still running

I'm retarded

laptop will be 4 years old next month, pic is for the hard drive that came with it. what do?

what does this mean? it's about to die? but I bought this hdd in april, that's only 7 months

>all that conflicting data

Quick rundown and ranking on best and most realiable maximum lifespan HDD brands?

Your disk killed 2.7MBs of data which you overwrote at some point.
Another 32KBs of data have died in the mean time, that your drive has just even managed to detect, data is already corrupt.

Use BRTFS if you value your shit.

I've had 2 Seagate drives before, 1 was backing up the other, both catastrophic failures after 3 years. And with statistics constantly putting Seagate at the bottom for harddrive life it started to make sense

>brandless
It's a Seagate.

It's not a seagate. I think it's buffalo or something, but i replaced the case

>10 years

But that's datacenter stuff, enterprise level hardware. Consumer grade HDDs are lucky to live past 4 years without bad blocks

shit

ST in the name stands for Seagate.

It's a seagate, any hard drive that is not under the name Hitachi, Toshiba, Samsung, Seagate or western digital 99% of the time is just a rebrand of those 5 brands

Replace the drive if any of these pop up:
SMART 5: Reallocated sectors count
SMART 187: Reported uncorrectable errors
SMART 188: Command timeout
SMART 197: Current pending sector count
SMART 198: Uncorrectable sector count
plus this one - End to end sector count or something like that. Airflow temp messages can be safely ignored if it's only temporary like (ex: your doing a full system image of it). Now if's constant message then your cooling factor is not good enough

Get an advanced filesystem you retards.

Convert the raw values to decimal you fucker. Thats the only thing thats valid.

>caution

I'm seeing many very young drives already in the "caution" or "bad" categories itt, what the fuck is going on?

seagate is bad

So besides the average Joe's Recuva... What data recovery softwares do shops usually run/use to salvage data from dead drives and data that's "unrecoverable"?

>seagate 2016
can you really trust them now?

Seagate is strong!

guys can i run this on windows and will it work even though half the drive is ext4?

>Subjecting to cucking yourself to SeaShit

Enjoy your dead drives within a year

>crystaldiskinfo showing more than four drives
how the fuck? at any rate, reallocated sectors count being that bad means that data backup will not be easy, I had a drive that was in the yellow for reallocated sectors and trying to backup data to a new one was an arduous process

you can

The drive has been strong since 2011 and i trust it to be still working in a few years.
But i'm not stupid I already have everything important backuped up twice

i have a 250gb seagate in one of my older servers. i got the hdd for like 20 usd from a guy on ebay some years ago. it came with bakery labels on it, was pretty shit looking. Runs for all these years fine, but warm. I hate seagate because i experienced many of their drives failing on me, i am a hgst use mostly for my storage servers. Was going with wd ae gold shit for a bit but they suggested i use their hgst company drives for what i wanted.

Uh, should I be getting worried? How long has this drive got? The other yellow drive is virtually identical (same model, near same stats)

In the past your drive killed ~1/4 a MB of data.
You at some point overwrote it.

It may misbehave again. I have no idea if it will go again, that depends on how it's feeling.

...

>seagate is bad
>less failure rates than WD as of 2016

Are you brain damaged?

How do you read that? how can I learn to read smart?

One:
Crystal diskmark is retarded.
It uses hex values so you can appreciate their pretty caution lights instead of getting the hard numbers and not know what you're looking at.

Gsmartcontrol uses decimal and has access to self test and previous temperature logs.
44 in hex > 66
66* 4KBs/ sector= 264KBs of data.

Two.
Put your shit on an advanced filesystem if it is valuable to you.

Data decays and windows has no means of safeguarding it.
Even backups can potentially push corrupt data which you later restore only to cry.

Change the raw reading to this, makes it easier to look at in my eyes.

Why are Toshiba so bad? I see lots of them fail too.

To do so, I pretty much need to buy new drives, convert them to an advanced file system, then move everything over to them, right?

You can easily change the defaults but I agree its silly to be defaulting to hex values when its people trying to read them.

Is the data really lost or just copied to another sector? I thought the "Uncorrectable Sector Count" was for the data that would be lost.

>Toshiba hard drive
>never seen one that didn't have something wrong with it in 20+ years of using them
>worst brand of hard drive by far
>1 bad sector in the OP image
>drive considered BAD
>no surprise

Stop buying Toshiba storage devices, people, seriously.

Data is always lost.

What occurs is:

Sector doesn't read
If consumer disk: (WD blue, green)
Retry endlessly and lock the running program until it comes back
If Enterprise disk: (WD black, red, most SSDs)
Retry for X number of seconds before telling the OS it's corrupt and crash running stuff

Something just sent a request to overwrite the bad sector so the disk breathes a sigh of relief and just reallocates it to spare, leaving the hole on the platter and using a spare area for new data.
The old mapping is permanently destroyed as it figures there is no need for it anymore.

Modern disks are fucked in terms of integrity.

Advanced filesystems are insanely flexible, ie BTRFS.

You can shrink your data partition and create a BTRFS partition and transfer your stuff bit by bit and keep expanding the BTRFS partition to cut more data over to it until fully migrated.

You can also use LZO or ZLIB compression to shrink your data and fit more on the same disk while writing and reading faster, assuming your CPU can keep up.

>BAD
First time I see this

Also, what do you call total drive failure?
because the first 3 months is mechanical shit,

My understanding of bad sectors is just 1 isn't shits going to die, abandon ship.

>Need to get more expensive enterprise drives, else data losses is your fault
Then may as well get SSDs at that point

All my Seagate's have more than 5 years
My old Samsung will make 10 next year.
I love SEAGATE so much that I just bought the new FIRECUDA as my boot drive, instead of some pleb-tier ssd pff

are toshiba pendrives ok?

I retired a WD 1TB drive the other day when it hit 70000 hrs, replaced it with a new 4TB drive, got 2 more WD 1TBs in the 60000 range, will replace and consolidate them when I get another 4TB+ drive.

Guess I really should do those backups I planned for today...

How bad is this? Sector count remains like this for a quite long time.

>pending sectors
lol just fucking scan and fix bad sectors already. it says you have potential bad sectors