You surely don't quick format your new hard drives, do you Jow Forums?

You surely don't quick format your new hard drives, do you Jow Forums?
>tfw waiting an hour now
Am I retarded?

Attached: c47864403e25c9115fd7ed5c21aa412b-d98wxnx.png (597x380, 159K)

Full format also scans for bad sectors so I guess it's not a bad idea

I let God sort them out.

>an hour

Jesus fuck you little babby, try Full formatting an 8TB drive then come back and complain.

Attached: 1538353678682.jpg (700x1050, 32K)

yes, this is retarded

full wipe is only needed for security when you are going to sell your drive

Its good for nuking old CP

I do quick formats only.

It even works on Windows, but pretty much all the Linux formats also are "quick format" (eg. doing a few spot checks and putting the initial metadata of the filesystem in place rather than doing a full wipe / wipe-test on every block).

There is not really any reason to do a full format, the only thing that I think really might happen is that if you loose your metadata (why?!) and you don't have a backup yet the data was important (why?!), maybe you pick up wrong things with the forensics tools that you use to try and piece back together some data.

I suggest if you MIGHT have a security issue when you're going to sell your drive, you probably better just encrypt it now.

>dd if=/dev/zero of=/dev/sdb && dd if=/dev/urandom of=/dev/db && dd if=/dev/zero of=/dev/sdb && dd if=/dev/urandom of=/dev/sdb

GOML

if you're worried about "security", don't sell your drive at all. Keep it or destroy it. If your security is worth less than the value of a hard drive, then you shouldn't be troubling yourself with worrying about what kind of drive wipe you're using.

>of=/dev/db
this is why you quadruple check before you run dd

The safest way to wipe a drive is to do a quick format and then set up full disk encryption (LUKS, not hardware since it's most likely backdoored). Since you're not going to have to retype the key ever again you can use a 1024-character key. Can get rid of the LUKS header as well if you want.

>using spinning rust in 2018
>not using tape backup

SSDs don't work like that.

Also possible. Although it's not TOO crazy to rely on encryption.
There's probably nobody targeting you or all discarded / sold drives that can break it.

If they *could* break currently considered safe encryption or hash functions, it'd possibly be a lot more fun to get at the bank's money and pull current data off everyone's computers... or just take down a few nations for general terroristic havoc.

>nuking old CP
fuck you

I don't get it, so is it good to quick format a new drive or not?

I do a full format. Then fill the drive 3 times, then do a deep scan. If it checks out, I keep using it. Takes about 3 days.

Haven't had a hard drive fail me that passed this test... ever. The first ones I ever used were 20 MB from 1984 and they still work as well.

> wait for an hour
go to sleep user, this should take days.

You are retarded. You need to surface test that fucking drive to check if they haven't sent you factory garbage.

The one time in years I didn't do a full write, test, h2testw, all that good shit on a new disk I bought (hard disk sentinel is a godsend btw) it died after three months and I spent hours recovering everything.
If I wasn't a retard and did a full format/test at first, I would have realised over half of the fucking sectors were bad, and got it replaced before I started writing to it.
Hard disk sentinel, or, if you're poor, h2testw (Or f3 or whatever)
HDS works great because you get a nice graph or write speeds, so you can tell how many sectors are bad, but not too bad that you get a failure. So you can tell if a "Working" drive that reports healthy is actually totally fucked

so this... is the power... of placebo.......

The format isn't finished you dipshit.

See A more measured approach would be
>dd if=/dev/random of=/dev/sdx
Then using hard disk sentinel to do a full write to the disk of a test pattern, reading back the pattern to confirm, and visually inspecting the read/write time graph it gives you to see if there are any bad clusters. If so repeat the process to see if they've been relocated, or simply moved (the latter means junk the disk)
Then format in your FS of choice, and run h2testw or F3 for a full write, then a full read.

If you wanna be really pedantic, write all 0's to the disk, then check the disk is fully zerod. Then write all one's, make sure it's all ones, then go back to zeros. This makes sure no bits are sticking.