VITAS changed the topic of #spacedock to: Problems?: https://github.com/KSP-SpaceDock/SpaceDock/issues | Matrix/Riot Chat: https://im.52k.de +spacedock:52k.de Feel free to ask for help, we only bite a little bit! | If you want to help, please check https://github.com/KSP-SpaceDock/SpaceDock-Backend/issues/5 :) | <VITAS> inet users have the attentionspan of a squirrel....oh a nut!
Webchat523 has joined #spacedock
Webchat523 has quit [Client Quit]
VITAS[m] has joined #spacedock
* VITAS[m] is travelign a lot atm.
VITAS[m] has quit [Quit: Idle timeout reached: 10800s]
Daz has joined #spacedock
VITAS[m] has joined #spacedock
<VITAS[m]> just fell into home :)
<VITAS[m]> HebaruSan: is code commited by AllisTauri and RockyTV i merge autodeployed onto alpha yet? (aka can i visit alpha.spacedock.info and see it in action?)
<VITAS[m]> btw AllisTauri we should do a feature freeze soon so we can get the code tested and promoted to beta. the alternative is to wait till the end of august when im back home.
<VITAS[m]> (yes im home now but only till the 10th)
HebaruSan[m] has joined #spacedock
<HebaruSan[m]> Deployed, yes; the alpha server is current. Auto-deployed, no; I had to deploy manually after the last merge.
AllisTauri[m] has joined #spacedock
<AllisTauri[m]> My commits, if they don't have bugs in them, should not be seen; no features in them.
<AllisTauri[m]> As for the freeze, I'd like to merge and test the new frontend building framework and docker environments I made.
<VITAS[m]> So well wait till the end of august. AllisTauri should have his stuff ready by then (by his rate of commits im sure of that) and HebaruSan can tinker with autodeploy if he likes (linuxgurugamer could help you once he figured out how to connect )
<HebaruSan[m]> Yeah help with auto deployment would be greatly appreciated, I'm at something of a stand-still due to lack of info about why it's failing.
<VITAS[m]> k ill give him vpn acc tomorrow because he cant get ipv6 to work on his end
<VITAS[m]> and ask him to join here
<AllisTauri[m]> After that I have a few ideas of how to better modularize the app. Including switching to webpack and separating celery related code.
<AllisTauri[m]> BTW, I need to look at deployment process so as not to break it with the next change.
<VITAS[m]> deployment as in what HebaruSan does?
<VITAS[m]> or as in the readme isntall howto?
<VITAS[m]> (that could use updating)
<AllisTauri[m]> Yes, both manually and auto
<VITAS[m]> i dint give you alpha cc yet or did i?
<AllisTauri[m]> I mean deployment on alpha and so on
<VITAS[m]> acc
<AllisTauri[m]> As far as I know, not yet
<VITAS[m]> k do i have a pub key from you?
<AllisTauri[m]> Yes
<AllisTauri[m]> Here in pm
<HebaruSan[m]> It's pretty primitive, it just tries to do a pull and then restart itself via systemd. There are probably much better ways, but I'm not well versed in hosting stuff like that.
<VITAS[m]> found it
<AllisTauri[m]> Then I'd like to include the `prepare.sh` into repository
<VITAS[m]> i added you as user "allista" with your key
<VITAS[m]> HebaruSan: can tell you how to connect
<VITAS[m]> what was prepare.sh for again?
* VITAS[m] just reads the query
<VITAS[m]> darklight did the systemd stuff
<VITAS[m]> i (until moving servers) ran gunicorn via bash script that in turn spawned abunch of screen sessions
<VITAS[m]> :/
<VITAS[m]> all the docker reminents in the source are from when docker was a new thing and we started playing with it.
<VITAS[m]> btw: darklight hates docker for some reason
<DasSkelett> But it's great for testing. A simple `docker-compose up`, go to localhost:5000 and there you have it. No annoying dependency management, or shooting the own local webserver by accident.
<HebaruSan[m]> It'd be nice to clean up the repo so there aren't unused files in there to confuse people
<VITAS[m]> yes tto both of you
<DasSkelett> But of course you don't have to use docker in production.
<VITAS[m]> i was on a train without inet for 7h
<VITAS[m]> so i had a bit of a thought
<VITAS[m]> if i can find some ddr3 ram to stuff in the machmini server ive kicking arround i could set up a dedicated kubernetes or docker host to play arround with it.
<VITAS[m]> (and i would conencted it to a decent inet con with fixed ip)
<VITAS[m]> problem beeing (atm at least) that ill be in inet wasteland starting from the 10th till the end of the month. (i will get inet at some point but doent know when)
<AllisTauri[m]> I've already removed most of the unused stuff and merged all the scripts into a single admin CLI, so there's that.
<AllisTauri[m]> *in a local branch
<VITAS[m]> cool
<VITAS[m]> i hope youre not running out of steam by the speed youre going
<AllisTauri[m]> Out of steam hardly, but from time to time I can run out of time with work and family.
<VITAS[m]> you know this is a marathon not a sprint :)
<AllisTauri[m]> Sure, but I wanted to make all the rearrangements before more active development starts to prevent most merge conflicts.
<VITAS[m]> im not holding you back and i surely hope that "more active development" will be a thing in the future (not only by you but others too)
<AllisTauri[m]> Besides, much of the changes so far are either technical, or use code from other projects, so I don't write from scratch.
<AllisTauri[m]> <irc_DasSkele "But it's great for testing. A si"> It would be `start-server.sh` now, as not everything may be set up via compose. Migration and population of the DB, for example.
<AllisTauri[m]> HebaruSan: so, how does one connect to alpha?
<HebaruSan[m]> I PM'd it to you on Discord, give me a moment to try the same via IRC...
HebaruSa- has joined #spacedock
<HebaruSan[m]> OK, sent. Did that get through?
<AllisTauri[m]> Ah, I see it on discord, thanks!
<AllisTauri[m]> Thanks, I'm in
<DasSkelett> @AllisTauri: I think I will have to see the changes itself to be able to judge anything. But as long as it stays as simple as a few commands without really manipulating the host system (outside the spacedock folder) it's probably fine.
<AllisTauri[m]> That it stays, yes)
<AllisTauri[m]> Yep, thanks HebaruSan , I'm already exploring there =]
<AllisTauri[m]> We currently don't use `uwsgi`, do we? So the `app.ini` config at the repo root is obsolete?
<VITAS[m]> i think we tried that at some point years ago
<AllisTauri[m]> ok; for docker I prefer to configure uwsgi right in compose config, so out with it
<VITAS[m]> just remember: alpha and beta are allready lxc container so im unsure if its possible or wise to run docker in them
<VITAS[m]> but for local dev envs its up to you
<AllisTauri[m]> Didn't intend to run docker there 🤣
<VITAS[m]> k :)
<AllisTauri[m]> I just make two docker environments: for development and for hypothetical production server that runs swarm or kuber or even plain compose. Just to be ready if we want/need to make such a change.
<VITAS[m]> ok thats good
<AllisTauri[m]> And I use `uwsgi` in production config simply because I'm used to it; same with nginx vs apache.
<AllisTauri[m]> Hm... I don't see any scripts that would update `/etc/systemd/system/spacedock*` files
<VITAS[m]> if we ever use docker in production i will still want it to be apache because i know that better :)
* VITAS[m] transforms into bedman. night
<AllisTauri[m]> I tried that once... it's a pain; nginx images are really easy to use and configs are simple and small. The apache container I made once to imitate the production environment (SLES12) took me two days work >_<'
<AllisTauri[m]> * I tried that once... it's a pain; nginx images are really easy to use and configs are simple and small. The apache container I made once to immitate the production environment (SLES12) took me two days work >_<'
<AllisTauri[m]> * I tried that once... it's a pain; nginx images are really easy to use and configs are simple and small. The apache container I made once to immitate the production environment (SLES12) took me two days work >_<'
<AllisTauri[m]> night)
<RockyTV> VITAS[m], you asked if I wanted access to alpha/beta: I'd like a read-only access if possible, we should have people from various timezones with access to the server, so it's not a problem for one of them to fix it. I don't know how the server code works so I don't want to risk accidentally nuking the alpha/beta servers :P
<RockyTV> AllisTauri[m], you said something about caching static resources, we could adopt this idea from SirCmpwn/ddevault's sr.ht: https://git.sr.ht/~sircmpwn/core.sr.ht/tree/master/srht/flask.py#L321
<AllisTauri[m]> I meant cache the base URL for static resources as opposed to calculate it on each API call.
<AllisTauri[m]> * I meant to cache the base URL for static resources as opposed to calculate it on each API call.
<AllisTauri[m]> And it's here
<AllisTauri[m]> And I'm also off to bed ~_~
<AllisTauri[m]> night everyone
<RockyTV> 'night
HebaruSan[m] has quit [Quit: Idle timeout reached: 10800s]