Show Idle (>14 d.) Chans


← 2022-01-24 | 2022-01-26 →
awt[asciilifeform|billymg]: So I have a server with an external IP on my home net. I find that I can't access the server via the external ip from my home net, only via the internal ip. This is somewhat annoying. Is there a way to access (via ssh) the external ip from inside my home net that anyone is aware of?
asciilifeform: awt: pfsense's knob. may have to find equiv. for your particular system.
asciilifeform: generally tends to be gnarly, as req's support from your natron.
asciilifeform: (by default routers won't do this)
awt[asciilifeform|billymg]: asciilifeform: ty will investigate. just a pain to have to switch console tabs when I come back home/go out, etc.
asciilifeform: awt: a traditional headache w/ e.g. irc bouncers
billymg: made some progress with building the musl gnat. i followed ave1's instructions and ran the `./build-ada.sh </install/directory> > build.output 2>&1` successfully on my other, glibc based, machine
billymg: using the same 2016 bin from asciilifeform's mirror as the bootstrap
billymg: this produced what looks like a complete gnat directory, but i'm not sure what i'm supposed to do next (instructions unclear)
billymg: i tried copying that dir (the one produced by ./build-ada.sh) over to my musl-based system but they still suffer from bad links (e.g. output of ldd shows errors)
billymg: if anyone can help me out i have three main questions right now (to sanity check that i'm even attempting to do the right things) -- 1) should the 2016 gnat bin work on a musl system? 2) are ave1's instructions meant to be followed on a musl machine or glibc mac
billymg: i'm looking at this step in his instructions and wondering if perhaps this is what i want: "4) build x86_64 native compilers"
bitbot: (asciilifeform) 2021-09-03 asciilifeform: ftr finds the 'v depends on gnat' thing to be an epic barf, courtesy of mp & diana ; really oughta be corrected
billymg looks forward to signpost's republican OS
PeterL[billymg|asciilifeform]: the depends on gnat was just for the ksum, right? didn't somebody make a ksum in c at some point?
asciilifeform: PeterL: the jfw fella baked one in ansi c. haven't tried it tho iirc
billymg: ah, i might give that a shot then
PeterL[asciilifeform]: I vaguely remember trying it when he released it, but I am not sure
PeterL[asciilifeform]: note to self: keep better notes
billymg: PeterL: are you operating on a musl-based system currently?
whaack[asciilifeform]: i've used jfw's c ksum, worked great iirc
PeterL[billymg|asciilifeform]: no, don't have one at the moment
billymg: whaack: nice
billymg: whaack: do you also have gnat on your system? is it musl-based or glibc?
whaack[asciilifeform|billymg]: billymg: tbh i'm not sure, iirc though i had some old version of gnat that couldn't run some other keksum and that's why i went with jfw's tool
billymg: whaack: my past experiences with getting the republican build system to work have always been a series of stumbles, trying this and that, until it finally somehow works and then i just go with it and forget how i got there until i have to do it again
billymg: with this build i've been documenting the steps, so that i can hopefully compile it into a single guide, that even noobs can follow
whaack[asciilifeform]: billymg: awesome, that's the way to do it
whaack[asciilifeform]: problem is that often when you are doing something you don't know until after you've stumbled 5 times that it's something worth documenting
asciilifeform has a half-finished asm 'keksum', maybe in peacetime will finish..
awt[asciilifeform|billymg]: ugh reloading blatta into my brain has been a bitch.
asciilifeform: how's that
awt[asciilifeform|billymg]: one particular function is almost 100 lines long
awt[billymg]: loaded with conditional logic
awt[asciilifeform|billymg]: proving very difficult to take apart
asciilifeform: imho would muchly benefit from e.g. cl's clos
asciilifeform: (or similar sane abstraction kit)
awt[billymg]: it handles every type of message
awt[billymg|asciilifeform]: logic for direct and broadcast messages is substantially different and should be separate
awt[billymg]: same goes for getdata and ignore messages.
asciilifeform: awt: asciilifeform considering to make current draft into 0xfa, even tho strongly dissatisfied w/ the fork logic
asciilifeform: given asciilifeform's present inability to come up w/ a substantially improved one
asciilifeform: afaik fixed 100% of the obvious gaffes
asciilifeform: ( if not -- plox to write in )
TCL-A2[asciilifeform]: hai shinohai
TCL-A2[billymg|asciilifeform]: testing blatta from Android phone weechat.
TCL-A2[billymg|asciilifeform]: Linux localhost 4.9.190+ #1 SMP PREEMPT Thu Oct 21 09:35:33 CST 2021 armv7l Android
← 2022-01-24 | 2022-01-26 →