How many is too many?

Nothing wrong with raid long as you use it as it's intended: Creation of large volumes w/protection against sudden failure. It gives you time to backup the data before the whole volume is lost for good. Key word here is "Backup" You still gotta have them. Aside note: You can run multiple raid arrays on a server. Advantage is shorter raid build times, data is split up so its not all in a single failure point, side step the 10-12TB unrecoverable read error thing, shorter time required to backup.

I think thats enough op
no more no less

Just delete everything every second year or so. Video formats and resolutions become obsolete after time and don't tell me you still use those reaction images you saved years ago.

>and don't tell me you still use those reaction images you saved years ago
Those aren't taking up space. The massive games, emulation and animu folders are.

well i have around 15+

This. I got fed up with external drives, bought a cheap ATX case and now I've got twelve drive bays to play with.

Just don't use RAID 0. The entire point is reliability and redundancy.

:^)

Three is too many.
You only need a raided pair at all times

Get ready for proof of storage m8y (^:

No u :^)))))))))))))

Attached: 1522452965939.jpg (4011x2100, 3.73M)

3

Attached: 72D24F0A-B8D1-43E2-9F0F-92DA1CBEFFD5.gif (295x235, 765K)

That's because people use RAID as backup solution.

My current setup

Attached: Screenshot_20180407-213411_Remote Desktop.jpg (1617x1306, 201K)

Raid6 or a zfs pool bb

Due to the growth of disk capacity without corresponding increase of (specified) disk reliability, in 2007 Robin Harris wrote an 'article' [zdnet.com/article/why-raid-5-stops-working-in-2009/] on the (purely theoretical) assumption that a RAID5 will not provide reliability, because the probability to get a failure of a second disk - or more precisely Unrecoverable Read Error (URE) - during a rebuild will be too high. LOTS of people have cargo-culted into this belief - many swear blind that it's happened to them (magically, they only started claiming this AFTER the article was published), they rattle off the URE specs of hard drives and nod their heads knowingly, they kick and scream if they see even home users deploying RAID5s, etc. While it appears statistically sound, the problem is they're all wrong, for a few reasons: they confuse full disk failures with inability to read one sector; these arguments are based on specifications of reliability given for bitstreams rather than for block devices; and finally, data about disk reliability declared by vendors is conservative. For example, vendors say that the you'll hit a URE after around 10^14 bits (which is about 12 terabytes). There are single drives that big in 2018 - so if they actually HAD the reliability specified by vendors, you wouldn't be able read the whole disk back without encountering an error. The cynic will see this for what it is: an attempt to scare people off cheap, ubiquitous RAID5 - built right into many operating systems and computer motherboards - onto more expensive (in infrastructure cost, byte-per-buck... and of course, consultancy cost) hardware mirroring and 'cloud backup' systems; it should be pointed out here that Internet connections have VASTLY higher data error rates than hard disks. Very few people - including the cultists who claim to have seen it - have ever been close to a storage system big enough for UREs to be any sort of day-to-day concern. (cont'd)

No, the main causes of data loss in a RAID5 are either disk failures due to common cause (so called 'common mode failure', such as poor environmental conditions, dodgy power, or bugs in the firmware - hence the market for 'NAS drives' that have firmware tailored to this workload), insufficiently quick disk replacement (one has to replace a failed disk in a RAID5 in a day or two, rather than putting it off for months - or until another drive fails), or someone screwed up and replaced the wrong disk - which then trashes the whole array when it comes online to rebuild. The latter can happen very easily: often your operating system's, computer's, or RAID card's idea of where drives 'are' doesn't gel well with the physical hard disks/ports inside the machine - and of course, very few of these IT 'gurus' are going to admit they made such a fundamental (though easy) mistake. Bottom line is, because of the high probability of common mode failure - which cannot really be engineered out of the system - RAID is not a replacement for backup. And this is the REAL problem here: 20 years ago, if your drive failed, you might lose a gigabyte of data - most of which was your bloated operating system and office applications, which are easy to replace. Now, you can lose multiple terabytes of data, which is all yours and often irreplaceable, and lazy people (and IT guys) are looking for a panacea - real or imagined - to stop them having to think about it. Robin Harris knows all this - he's a snake-oil salesman, absolutely pulling himself over how many people he's taken in - and that's why he followed it up with a hit piece against RAID6 [zdnet.com/article/why-raid-6-stops-working-in-2019/]. And guess what? Sponsored ad links on that article include 'Cloud Storage', 'RAID Controller Card', and 'RAID 10' - all of which deliver far less bang-for-buck, and pays Robin for the click-through. Draw your own conclusions.

I don't get it, does it mean the whole server is fucked because of one drive ?

yeah
the guy put 30TB in a RAID0
so basically the guy lost 30TB of data

one for your backup
one for your backup's backup

>2
>3-4
>6
way to logic, idiots

well they are pretty cute desu

Jow Forums in general knows slightly more than jack and shit about hard drives, beyond "get an SSD it's rly fast" "lol u don't even need more than a 1TB hard drive for gaymz it's current year" "bro don't RAID 0 it's bad"

there are exceptions of course, and not saying i'm an expert but in general .. well there is no storage / data hoarder general on Jow Forums (maybe there could be)

if you want to know or discuss NAS and data hoarding, there are much better places

Are externals not shit anymore?

Back in college about 7 years ago i had twoo 500hb and a 1tb hd all shit themselves in less than a year. The 1tb lasted 8 months and my first 500 shat itself in 3 weeks. Lost so much porn and anime.

Im only considering it because my laptop doesnt have a second slot for an hdd

Use a UPS if you run Raid. Helps eliminate "Write Hole" ex: you write data to raid 1 array, main power cuts out, that data is gone/damaged. plus upon power restore you gotta wait while the array rebuilds itself. UPS: Write data, power fails, server keeps right on going, data is safely written, no rebuilds, main power comes back. Also upon power fail, the ups will allow server will shutdown properly avoiding all the data corruption/rebuild issues.

The sealed ones are still shit. Get a laptop pull HDD and stick it in a USB 3.0 enclosure. People buy laptops and just pull the HDDs immediately to replace them with SSDs, so they're basically brand new. Check craigslist, kijiji, etc.

Seems like some really bad luck. I've had a 1TB WD MyBook since 2010 that's still fine.