# I am the Watcher. I am your guide through this vast new twtiverse.
# 
# Usage:
#     https://watcher.sour.is/api/plain/users              View list of users and latest twt date.
#     https://watcher.sour.is/api/plain/twt                View all twts.
#     https://watcher.sour.is/api/plain/mentions?uri=:uri  View all mentions for uri.
#     https://watcher.sour.is/api/plain/conv/:hash         View all twts for a conversation subject.
# 
# Options:
#     uri     Filter to show a specific users twts.
#     offset  Start index for quey.
#     limit   Count of items to return (going back in time).
# 
# twt range = 1 19
# self = https://watcher.sour.is/conv/6zzyulq
One of the important lessons:

I like to put as little strain as possible on the floppy disks that I have, especially when installing operating systems. I thus like to prepare disk images on my modern Linux box in QEMU (where I can use floppy images instead of actual disks) and then transfer them over to my real retro box.

Older operating systems like OS/2 make extensive use of CHS addressing and even store some of this information in the HPFS filesystem header. CHS info spreads all over the place. So, simply creating a QEMU disk image, installing something and then copying to another drive *probably* won't work, because QEMU guesses *some* CHS geometry that won’t necessarily match that of the target drive.

The solution is to a) create a QEMU disk image of the exact same size (in bytes) as the intended target drive, b) configure a matching CHS geometry in QEMU. The latter can be done like so:


-drive file=warp3.raw,if=none,id=disk1,format=raw
-device ide-hd,drive=disk1,cyls=495,heads=16,secs=32,bios-chs-trans=none


How do you know the correct CHS geometry? Ask the BIOS of the target machine.

And then be very thankful that we don’t have to deal with this anymore today. 😂
One of the important lessons:

I like to put as little strain as possible on the floppy disks that I have, especially when installing operating systems. I thus like to prepare disk images on my modern Linux box in QEMU (where I can use floppy images instead of actual disks) and then transfer them over to my real retro box.

Older operating systems like OS/2 make extensive use of CHS addressing and even store some of this information in the HPFS filesystem header. CHS info spreads all over the place. So, simply creating a QEMU disk image, installing something and then copying to another drive *probably* won't work, because QEMU guesses *some* CHS geometry that won’t necessarily match that of the target drive.

The solution is to a) create a QEMU disk image of the exact same size (in bytes) as the intended target drive, b) configure a matching CHS geometry in QEMU. The latter can be done like so:


-drive file=warp3.raw,if=none,id=disk1,format=raw
-device ide-hd,drive=disk1,cyls=495,heads=16,secs=32,bios-chs-trans=none


How do you know the correct CHS geometry? Ask the BIOS of the target machine.

And then be very thankful that we don’t have to deal with this anymore today. 😂
One of the important lessons:

I like to put as little strain as possible on the floppy disks that I have, especially when installing operating systems. I thus like to prepare disk images on my modern Linux box in QEMU (where I can use floppy images instead of actual disks) and then transfer them over to my real retro box.

Older operating systems like OS/2 make extensive use of CHS addressing and even store some of this information in the HPFS filesystem header. CHS info spreads all over the place. So, simply creating a QEMU disk image, installing something and then copying to another drive *probably* won't work, because QEMU guesses *some* CHS geometry that won’t necessarily match that of the target drive.

The solution is to a) create a QEMU disk image of the exact same size (in bytes) as the intended target drive, b) configure a matching CHS geometry in QEMU. The latter can be done like so:


-drive file=warp3.raw,if=none,id=disk1,format=raw
-device ide-hd,drive=disk1,cyls=495,heads=16,secs=32,bios-chs-trans=none


How do you know the correct CHS geometry? Ask the BIOS of the target machine.

And then be very thankful that we don’t have to deal with this anymore today. 😂
One of the important lessons:

I like to put as little strain as possible on the floppy disks that I have, especially when installing operating systems. I thus like to prepare disk images on my modern Linux box in QEMU (where I can use floppy images instead of actual disks) and then transfer them over to my real retro box.

Older operating systems like OS/2 make extensive use of CHS addressing and even store some of this information in the HPFS filesystem header. CHS info spreads all over the place. So, simply creating a QEMU disk image, installing something and then copying to another drive *probably* won't work, because QEMU guesses *some* CHS geometry that won’t necessarily match that of the target drive.

The solution is to a) create a QEMU disk image of the exact same size (in bytes) as the intended target drive, b) configure a matching CHS geometry in QEMU. The latter can be done like so:


-drive file=warp3.raw,if=none,id=disk1,format=raw
-device ide-hd,drive=disk1,cyls=495,heads=16,secs=32,bios-chs-trans=none


How do you know the correct CHS geometry? Ask the BIOS of the target machine.

And then be very thankful that we don’t have to deal with this anymore today. 😂
Hahaha, @movq:

> And then be very thankful that we don’t have to deal with this anymore today. 😂

That's exactly what I thought when reading your explanations. :-D
@lyse Yeah, it’s really, really annoying. 😂

I would have loved to transfer the contents of one particular hard drive onto a Compact Flash card, doing a 1:1 copy using dd – but that just won’t work. The card has a different CHS geometry than the HDD. I actually spent a couple of days trying to work around this: Reading/understanding/reverse-engineering OS/2’s boot loader code and trying to fix the incorrect bytes. It does indeed boot now and I learned a lot. QEMU is quite powerful and allows you to attach a gdb process to the machine, so you can single-step the instructions, read memory, and what not. But fdisk still shows errors, so I don’t trust it … Maybe writing to a particular area of the filesystem will crash the whole thing. 🫤

It’s a strange hobby that I picked there. 😂
@lyse Yeah, it’s really, really annoying. 😂

I would have loved to transfer the contents of one particular hard drive onto a Compact Flash card, doing a 1:1 copy using dd – but that just won’t work. The card has a different CHS geometry than the HDD. I actually spent a couple of days trying to work around this: Reading/understanding/reverse-engineering OS/2’s boot loader code and trying to fix the incorrect bytes. It does indeed boot now and I learned a lot. QEMU is quite powerful and allows you to attach a gdb process to the machine, so you can single-step the instructions, read memory, and what not. But fdisk still shows errors, so I don’t trust it … Maybe writing to a particular area of the filesystem will crash the whole thing. 🫤

It’s a strange hobby that I picked there. 😂
@lyse Yeah, it’s really, really annoying. 😂

I would have loved to transfer the contents of one particular hard drive onto a Compact Flash card, doing a 1:1 copy using dd – but that just won’t work. The card has a different CHS geometry than the HDD. I actually spent a couple of days trying to work around this: Reading/understanding/reverse-engineering OS/2’s boot loader code and trying to fix the incorrect bytes. It does indeed boot now and I learned a lot. QEMU is quite powerful and allows you to attach a gdb process to the machine, so you can single-step the instructions, read memory, and what not. But fdisk still shows errors, so I don’t trust it … Maybe writing to a particular area of the filesystem will crash the whole thing. 🫤

It’s a strange hobby that I picked there. 😂
@lyse Yeah, it’s really, really annoying. 😂

I would have loved to transfer the contents of one particular hard drive onto a Compact Flash card, doing a 1:1 copy using dd – but that just won’t work. The card has a different CHS geometry than the HDD. I actually spent a couple of days trying to work around this: Reading/understanding/reverse-engineering OS/2’s boot loader code and trying to fix the incorrect bytes. It does indeed boot now and I learned a lot. QEMU is quite powerful and allows you to attach a gdb process to the machine, so you can single-step the instructions, read memory, and what not. But fdisk still shows errors, so I don’t trust it … Maybe writing to a particular area of the filesystem will crash the whole thing. 🫤

It’s a strange hobby that I picked there. 😂
@movq Oh wow, couple of days!? You're *very* passionate about that. ;-) Respect, reverse-engineering this sort of thing is waaay beyond my capabilities. This sort of stuff is too low-level for me.
@lyse Well, it’s one of those cases where you keep thinking “god dammit, how hard can it be?!” 😂 Quite hard, apparently, but I didn’t know that when I started. 😂
@lyse Well, it’s one of those cases where you keep thinking “god dammit, how hard can it be?!” 😂 Quite hard, apparently, but I didn’t know that when I started. 😂
@lyse Well, it’s one of those cases where you keep thinking “god dammit, how hard can it be?!” 😂 Quite hard, apparently, but I didn’t know that when I started. 😂
@lyse Well, it’s one of those cases where you keep thinking “god dammit, how hard can it be?!” 😂 Quite hard, apparently, but I didn’t know that when I started. 😂
@movq And all of a sudden, a week has passed. :-D Hahaha, I know. Let's *quickly* finish this tiny thing here…
@lyse … exactly. %)
@lyse … exactly. %)
@lyse … exactly. %)
@lyse … exactly. %)