@eyedeekay
&Irc2PGuest88200
&zzz
+R4SAS
+RN
+StormyCloud
+T3s|4
+acetone
+dr|z3d
+eche|off
+hagen
+mareki2p
+orignal
+postman
+radakayot
+segfault
+snex
+wodencafe
Arch2
Danny
DeltaOreo
FreeB
FreefallHeavens
Irc2PGuest53192
Irc2PGuest59134
Irc2PGuest59581
Irc2PGuest64993
Irc2PGuest96449
Onn4l7h
Onn4|7h
Sisyphus
Sleepy
T3s|4_
aeiou
ardu
b3t4f4c3___
b4dab00m
bak83_
boonst
cumlord
death
dr4wd3
duck
enoxa
eyedeekay_bnc
not_bob_afk
onon_
phil1
phobos_
pisslord
poriori_
profetikla
qend-irc2p
rapidash
shiver_1
solidx66
u5657
uop23ip
w8rabbit
x74a6
eyedeekay
Hm. Might have to think of an interesting recommendation to do with runners... the red X's should go away with the update I just made to the gitea server, but runners are going to be slow unless I put them on some other hardware, which I have, will just need to get it set up
eyedeekay
But in theory we could all just run whatever runners we want on whatever hardware we want and the runner would talk back to our gitea instance
dr|z3d
github or gitlab.com's CI is easy enough, just pull the artifacts once they've run.
dr|z3d
and you can probably hook into the process with gitea.
eyedeekay
Hm I hadn't considered using gitlab's CI, but we're already using github's CI, gitea CI and github CI are behaviorally identical
eyedeekay
gitea's CI is done using this `act_runner` program which will, for all intents and purposes, exactly follow the github CI
eyedeekay
But the neat thing is that you just grab this token from the repo you want to build for, and use it in the act_runner command on whatever machine you want to build with
dr|z3d
ideal
eyedeekay
So I haven't got a ton of space for runners on the machine where the gitea server is, but I have all kinds of space for runners on other machines
eyedeekay
I always thought the coolest way to do "github pages for I2P" would be run the action that generates the page in `act` and dump the output to eepsite/docroot
dr|z3d
or you could just forward requests for foo.github.io via nginx.
dr|z3d
and host as foo.i2p
eyedeekay
You can actually just stick it directly into the Hidden Services Manager if you don't want to have a local copy
dr|z3d
maybe, if you're only hosting a single page on github, or you've made sure all your links are relative.
eyedeekay
I actually think I have a guide for that...
eyedeekay
Oh yeah if you don't strictly use relative links you do have to rewrite stuff
StormyCloud
You need some more servers spun up?
eyedeekay
Maybe, could certainly help, but I could keep your costs down by using some of the boxes I have that are doing nothing but hosting terabytes of linux ISOs
zzz
I need a nonblocking I2CP lookup API for zzzot
zzz
it's killing me, I'm going to have to do a whole thread pool
zzz
seeing multiple bad inbound NTCP2 message 1 errors from two IPs 52.60.61.249 110.137.220.12; first one is AWS, I doubt either one is rust or go
zzz
started last night
altonen
i've seen a bunch of invalid/missing net id and a few invalid sig on RI for ntcp2 for a while now
altonen
maybe somebody is working on a new router and testing in mainnet
zzz
most of the weird stuff is i2pd routers badly modded by botnet people for unknown reasons
zzz
used to see i2pd w/o netid, not any more, I think it was an ancient bug at startup
zzz
but I may be blocking them before I get there now, just checked my logs, no wrong netid in last 2 weeks
altonen
ok fair enough
zzz
so just because I'm not seeing it doesn't mean it's not real
zzz
thinking if I should ban the IP after a corrupt ntcp2 session request but that might be a little much
dr|z3d
maybe count the # of requests in a short period, and if you keep seeing corrupt requests, ban hammer?
zzz
yeah but I don't want to open up some MitM attack vector where he corrupts packets
dr|z3d
ok
zzz
maybe we could ban for a few minutes but not a hammer. unfortunately we don't have a temporary IP ban facility now so it's harder
dr|z3d
we don't?
dr|z3d
not sure I got that missive
zzz
no, not for IP blocklist, only for hash banlist
zzz
there may be something low-level in the transports, I'm looking...
zzz
there was no missive, but there's also no API in Blocklist.java with a specified expiration, and you won't see IP expirations on /configpeers either
zzz
I'm getting about 3/hour total from those two IPs, for the last 18 hours
zzz
not a lot, but it could be a vector for a CPU exhaustion attack if we're doing DH on every garbage that comes in
zzz
but all our incoming is queued and throttled so I think we're safe
zzz
we do have a sort of temp IP ban in SSU2 but not NTCP2
dr|z3d
ok. not worth the effort adding an ip ban feature to the banlist?
zzz
banlist and blocklist are separate things
zzz
best to keep it down in the transports
dr|z3d
ok
zzz
I'll probably adapt the ssu2 thing for ntcp2
zzz
oh wait, we do have something in there
zzz
super, this will be a oneliner
dr|z3d
*thumbs up*
zzz
and found some more old NTCP 1 stuff while I was poking around
zzz
improved version of jetty webserver logo for the console, previously scoffed at by drz in saltr: cake.i2p/file/koBl3ZoUnc_xD9SgH4bMl8EdTeyiWeUNlf0dkMTeO_j219jSdzdA/jetty.png
zzz
proposed to replace the hamburger menu looking icon for the 'web server x' lines in the sidebar
zzz
basically a :// in orange
dr|z3d
LOL
dr|z3d
*** carries on scoffing. *scoff scoff* ***