I'm hosting a website on a RAID0 of 30 floppy drives(totallynormalwebsite.ddns.net) |
I'm hosting a website on a RAID0 of 30 floppy drives(totallynormalwebsite.ddns.net) |
Also-
Tell me you're into Chiodos without telling me you're into Chiodos.
Big ol dose of nostalgia listening to your old metalcore tracks.
> Server uptime: 20 seconds
> Server load: 157.41 158.15 155.66
oh
P.S I am not going to and I am drunk.
basically system memory size vs block device size was the opposite way around of today... then again people could and did have many individual floppy disks.
I can imagine many programs doing this though.
Every time you reboot, it will repair the array.
edit: thinking about it, there's some good educational value, but maybe via vm than plopped on the interwebs :D
- Put a bird^H^H ZFS on it
- Switch to RAID10 (a stripe of mirrors), and go 2/3 floppys wide so you can have some redundancy in each mirror gropu
- Get some Pis (or other SBCs) and hook those up and run Ceph... if this keeps going we'll have a SAN soon enough.
- ZIP disks?[0]
Also, I don't think I ever want to hear of the "hug of death" for any site ever again -- I don't think this site hosted on 30 floppies was hugged to death.
Though this may have been caused by me being in school and having to buy crappy white label ones. Couldn't afford those fancy imation ones with all the games I copied lol.
Thankfully, all entrance doors had an earplug dispenser...
Nothing is simple :D
Anyone want to try RAID0 on QIC tapes?
This is what I get, when I download the full image and convert to PNG: https://i.imgur.com/JF4wtMg.png
During the conversion (with imagemagick), I get these errors:
convert: Corrupt JPEG data: premature end of data segment `megafloppy.jpg' @ warning/jpeg.c/JPEGWarningHandler/403.
convert: Corrupt JPEG data: found marker 0x74 instead of RST1 `megafloppy.jpg' @ warning/jpeg.c/JPEGWarningHandler/403.
convert: Corrupt JPEG data: 206 extraneous bytes before marker 0xfb `megafloppy.jpg' @ warning/jpeg.c/JPEGWarningHandler/403.
convert: Corrupt JPEG data: found marker 0xfb instead of RST2 `megafloppy.jpg' @ warning/jpeg.c/JPEGWarningHandler/403.
convert: Unsupported marker type 0xfb `megafloppy.jpg' @ warning/jpeg.c/JPEGErrorHandler/345.
Looking closer, all bytes between offset 0x115C00 and 0x11F800 have been set to 0xf6, and all bytes from there until 0x11FC00 have been set to 0.Bytes from 0x2EFC00 to 0x2F5C00 have been set to 0, followed by 0xf6's all the way until 0x2FFC00.
I'd be curious to know what failure mode(s) conjured the 0xf6's into existence.
Edit: Original version is here https://web.archive.org/web/20220715175852if_/http://totally...
> I'd be curious to know what failure mode(s) conjured the 0xf6's into existence.
Today's fun fact: The MS-DOS `format` command fills the disk with 0xf6, not 0x00. Though this is linux running on Mac hardware, reading a disk that should have actual data, so maybe that isn't the reason.
Oh well. The irony is that to this day I have a tick of idly running `sync` at the command prompt, which I developed dealing with floppy and hard disk corruption running early versions of Linux. A crash or (IIRC) even a simple reboot sometimes resulted in disk corruption preventing Linux from booting. Reinstalling Slackware from floppy disks took quite awhile on its own, especially if installing the X11 disk sets, but half the time at least one of the disks would be corrupted, requiring me to download a fresh copy (using Windows--I was dual booting) over my 2400 baud modem, and then restarting the install from scratch. I probably went through this procedure at least a half dozen times, or at least enough to develop the tick. It was the best of times, it was the worst of times.... =)
In this case, the theoretical maximum bandwidth is 24MBit/s.
The problem is the old, slow usb bottleneck. I'm not sure how much faster, probably hundreds of bps rather than under 24, but a faster RAID0 rig would be to instead have 30x Mac G4 Digital Audios connected via gigabit switch, and share then RAID0 the internal floppies. It would also have whatever advantage running an XGrid PPC cluster on Tiger might provide. These boxes also ran PPC Ubuntu; no doubt Linux would eek out a dozen or so more bps, plus beowulf.
Usually it's way, way more: https://store.supermicro.com/server-accessories/fans.html?fa...
Oof.
Something like voice encoded at 32K sounded at least as good as a phone and played back off a 1.44 floppy and IIRC that was about the best that could be done.
You will probably be surprised how long an audio recording can be, if its voice at a low rate on one floppy. If you go variable bit rate and silence detection I subjectively remember "ten minutes" was quite reasonable on a 1.44 disk.
Extrapolating from historical experience, thirty or so in parallel should push over half a meg/sec quite reliably.
If you record speech onto a floppy drive off a cheap mic you'll record the sound of the floppy in the recording, which is funny to me.
I wish I still had those files. Useless, of course, but would be funny.
He got results ranging from 25 KB/s - 100 KB/s
As another example You could easily make RAM disks via a desktop menu, to facilitate the common setup of a single built in floppy drive. So you could copy files off into RAM via the GEM desktop GUI (literally drag and drop) and then put another disk in etc.
Random aside: I accidentally found that this machine would transmit audio over FM radio for a short but very useful range automatically.. blew my mind as a kid. I could never find anything official about this online later so not sure if it was an intentional design, poor EM design of the audio chip, or a hack of my particular 2nd hand franken-tari - either way it was super useful to have wireless audio in the 90s.
The ”FM” officially stands for Floppy & (RF) Modulator, but that’s some coincidence that the audio chip emits frequency modulated radio waves at a listenable frequency.
Star Wars theme: https://youtu.be/3KS02q0BUnY
I Wat to Break Free: https://youtu.be/lbd06i9B2wU
Luckily, the gen 7 tapes are 10-100 times that large, depending on who you ask. Each disk costs approximately $2-$5 as well, assuming you get some sort of discount. Tapes are more, they hold more, but at a certain point having a disc in a jewel case is better than having some tape. bluray drives are easy to find, tape readers, not so much.
I don’t think those are actually several TB. They’re fakes that are advertised as having several TB.
Portable flash drives using M.2 drives internally could reasonably be up to 8TB, but being at the high end they're expensive.
If i hadn't seen it myself i would have wondered why it wasn't already possible already.
sync;sync;halt was once a legit way to shut down ;)
Another fun fact: That character is ö under Windows-1252, the codepage in use by Windows since about Windows 2.0, and some late MS-DOS.
It seems necessary that the blocks storing the directory entries are cached too. Otherwise every non-existent lookup (that doesn't hit a negative dentry) would hit disk, each separately.
So yeah, the page cache is keyed by file, but the system should still cache the directory structure.
The DDoS scenario would be doing GET requests of random nonexistent filenames. You could change the name at every step so that the check for a negative dentry is never a cache hit.
Also the page/buffer cache would probably cache the FAT direntries so probably not a huge issue here.