Appears I have way too many Linux ISOs. What do people whose storage space is filled with Linux ISOs do?

Correct, they get more storage.

My home server consists of a computer whose creator specifically designed it to extort currency from clients. This means I am unable to upgrade its measly 256gb internal storage. For a while I used a random NVMe to USB enclosure to extend my storage space by another random 256gb.

It would’ve been fine and I probably would’ve kept it this way a little longer if it weren’t for the tiny scare a week ago.

I was in the midst of acquiring more Linux ISOs when The-Linux-ISO-Downloading-Application displayed the torrent state as “Errored”.

Not good.

But the disk shouldn’t be full yet? Wait where is the disk?

Not good.

I think the issue was that the external disk got too hot for it’s liking. I didn’t dig too much into the logs as this occurrence was significant enough to raise the Wife Approval Factor of getting better storage.

Btw, I know you’re asking “But Kaur, can’t you just re-download the Linux ISOs?”. I could, but the two components of A. possibility of people not seeding at the time of necessity, and B. The Descendant demanding round-the-clock access to the ISOs make me interested in storing them in a safe manner.

Solution

Get two of the largest hard drives I can get my hands on. Those two disks turned out to be whitelabel 6TB pieces from datablocks.dev (not affiliated).

Why only 6 TB? Lack of currency.

Why only two? Because my server machine has two USB-A ports. I should’ve gotten a third, replacement disk, but refer to previous answer.

I did think about building a new NAS box, but didn’t really like the idea too much as I already had ample compute at hand and only needed storage. A friend of mine swore by SATA-to-USB adapter thingies, so after some consideration and research into DAS enclosures, none of which turned out to be good, I settled on the idea of doing SATA over USB.

I eventually went with two of these Inateck FE3003 (not affiliated) boxes. Those got chosen because a single guy on Reddit (redlib, reddit) said ASMedia USB controller chips were the best, and that Inateck made enclosures with chips from said manufacturer.

The Inateck FE3003 things seem to sport the ASMedia AMS225. I don’t know if it is good or bad, I’ve not run any tests, and I’m solely basing the purchase on the recommendation from the guy above.

Seems both of my drives have two previous power-on hours. Second-hand? Yes please!

Next steps

First step is to read this blog post on how tf I am supposed to set the disks up.

In true internet fashion I’ve not actually read the post but found this command badblocks -wsv -b 8192 /dev/sdX and shall simply paste this into terminal. What’s the worst that could happen, right?

After badblocks is done I should look into setting up a RAID array. Since I’m going after redundancy and only have two drives, RAID 1 seems appropriate.

The author of the above post claimed they ran their badblocks for 24 hours. Quik mafs determined that my 3x smaller drives would take 3x less time. 43 hours later I am not convinced.

Chipset

After interrupting badblocks to get my damn drives back! I’ve stared creating partitions. Running sudo parted /dev/sdb gave the following output:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10

GNU Parted 3.6
Using /dev/sdb
Welcome to GNU Parted! Type 'help' to view a list of commands.
(parted) print                                                            
Error: /dev/sdb: unrecognised disk label
Model: Asmedia ASM1153E (scsi)                                            
Disk /dev/sdb: 6001GB
Sector size (logical/physical): 512B/4096B
Partition Table: unknown
Disk Flags:

Hmm… “ASM1153E” is not quite the promised “AMS225”… I’ll contact the seller to see what’s up, but I might not care enough to replace it.

For proof, here’s a WayBackMachine’d page from today (“today” is now long gone).

Wrong chipset

Inateck’s reply

I’ve been in contact with Inateck support about this, and they’ve confirmed the chipset is indeed ASM1153E. They offered a full refund if the enclosures are brand new and unused (which they are not) and a shipping label for me to return the enclosures, or a 10% refund if I decided to keep the product.

I decided I don’t care enough to return the enclosures and look for new ones. Partly because while the chip provides only 5 Gbps of bandwidth, SATA III caps out at 6 Gbps anyway and the drives are for bulk storage so the lack of a gbps will not be noticeable. But mainly because I don’t want to go through the installation (placing them on a shelf) and cable management again.

Current situation

It is now exactly 5 months after I started writing this post, and I’ve nothing interesting to report.

Well, after writing the above sentence I thought I’d run the SMART data test control thing command to make sure I indeed didn’t have anything interesting to report. Both drives are reported as passing, but they also say a bunch of attributes are of Old_age or Pre-fail state.

I’ve spent some time in my life around computers to know that such states do sound like the “ignore happily” category, so I shall now commence intense sweating before I go on a vendor-neutral-internet-searching spree to figure out if I have left one day or two days before all my Linux ISOs get shafted.

Toodaloo!