• 0 Posts
  • 12 Comments
Joined 11 months ago
cake
Cake day: September 30th, 2025

help-circle

  • I’m not sure how a firewall would help, you’d definately want to do this from the routing table.

    isnt parsing 1000+ subnets are load on cpu?

    IPv4 addresses are basically just 32bit numbers, any somewhat modern CPU can check through 1000 routes in nanoseconds.

    Also 1000+ seems a bit excessive, you should be able to aggregate some of those subnets into a few bigger ones. There are calculators online for this, or you can script it.



  • I actually made a similar thing a few years ago, and mostly relied on WebSockets with a HTTP fallback: https://github.com/Steve-Tech/My_Time

    You actually seem to know what you’re doing with Workers though, mine was pretty much a half-assed attempt at porting my python code (which again could probably be better).

    Also my desktop is PTP synced to a GNSS disciplined time server, and I’ve found the My_Time workers demo to be fairly accurate. I also feel Cloudflare should be reasonably accurate given they have database products you’re intended to use with workers, but yeah, there’s no guarantees.





  • The os is a layer that mediates these devices.

    The OS doesn’t just mediate the devices, it also provides a consistent interface for software to talk to the hardware. E.g. software doesn’t care if you’re using a USB or PS/2 keyboard, the operating system handles that.

    I’ve seen some interest in bare metal web servers

    Usually in the context of servers, bare metal means it’s not running in a VM, and you are dedicated to the hardware. E.g. one server may otherwise be running multiple customers all isolated from each other using VMs, with bare metal servers you are the only customer using the hardware. They’re supposedly more secure as there isn’t another customer that could use some VM escape vulnerability and read your data. It’s nothing to do with whether you are running an OS or not (although no OS is very not practical on production servers).


  • If you can’t find an easier way, compiling your own kernel isn’t too hard. After you’ve git cloned the new kernel, you can just copy your distro’s kernel config (they’re usually in /boot), and then use make deb-pkg -j$(nproc) to compile to a .deb file, so it’s easier to uninstall.

    At the start of compilation it will ask about a few new options, you can just press enter to use the default option, or decide for yourself. Probably don’t worry about make menuconfig unless you want to.

    There’s proper tutorials online if you wish to do this.