IRC logs for #openttd on OFTC at 2021-05-11
⏴ go to previous day
00:02:18 *** iSoSyS has quit IRC (Remote host closed the connection)
01:54:15 *** snail_UES_ has joined #openttd
02:45:18 *** Wormnest has quit IRC (Quit: Leaving)
02:53:18 *** supermop_Home has quit IRC (Ping timeout: 480 seconds)
02:53:52 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
03:46:40 *** lobstarooo has joined #openttd
03:51:49 *** lobster has quit IRC (Ping timeout: 480 seconds)
03:51:51 *** lobstarooo is now known as lobster
05:49:10 *** snail_UES_ has quit IRC (Quit: snail_UES_)
05:57:12 *** HerzogDeXtEr has joined #openttd
06:02:23 *** andythenorth has joined #openttd
06:05:15 *** nielsm has quit IRC (Ping timeout: 480 seconds)
06:07:56 *** sla_ro|master has joined #openttd
06:08:57 <DorpsGek> [OpenTTD/OpenTTD] fearlezz77 opened issue #9237: I never get the new trains after 1980- but i get all plains, boats and cars https://git.io/JseBp
07:16:04 *** bruvvha has quit IRC (Read error: Connection reset by peer)
07:16:29 *** bruvvha has joined #openttd
07:24:22 *** y2kboy23 has quit IRC (Ping timeout: 480 seconds)
07:32:38 <TrueBrain> right, time to look into the networking issues we had yesterday ...
07:32:43 <TrueBrain> no clue how to start :P
07:43:06 <Xaroth> Well yeah, could've been Laibach or something weird like that.
07:46:52 *** y2kboy23 has joined #openttd
08:05:22 *** y2kboy23 has quit IRC (Ping timeout: 480 seconds)
08:08:24 *** andythenorth has quit IRC (Quit: andythenorth)
08:17:49 *** Speedy` has quit IRC (Read error: Connection reset by peer)
08:19:00 <TrueBrain> wow, our "when to tell clients someone left" code flow has become rather complex
08:19:05 <TrueBrain> with all kind of assumptions etc
08:21:14 <TrueBrain> and really easy to reproduce
08:27:47 *** andythenorth has joined #openttd
08:28:11 <TrueBrain> I cannot believe this hasn't been reported more .. I need to check the source of 1.11 ..
08:28:23 <TrueBrain> but from what I read, any client-crash will never result in a leave message
08:28:36 <TrueBrain> guess that doesn't happen often enough to be noticed?
08:29:04 *** andythenorth has quit IRC ()
08:30:09 <TrueBrain> no client is notified yet (also not its job)
08:30:30 <TrueBrain> which explicitly ignores either of the two values of the former, and does not tell anyone about the leaving of the client
08:44:12 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9238: Fix: [Network] clients leaving because of broken connections was not broadcasted https://git.io/JseAw
08:46:44 <TrueBrain> I cannot believe nobody reported it, really
08:53:41 *** y2kboy23 has joined #openttd
09:01:32 *** sla_ro|master has quit IRC ()
09:01:38 <TrueBrain> sometimes you find lovely bugs ..
09:01:48 <TrueBrain> opening the network list, means _network_host_list is added to the list
09:01:52 <TrueBrain> this is done via NetworkAddServer
09:01:58 <TrueBrain> which rebuilds the _network_host_list
09:02:08 <TrueBrain> yeah ... that is not going well
09:08:37 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9239: Fix: [Network] don't mark the last-joined server as manual https://git.io/Jsvvn
09:09:50 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9240: Fix: [Network] don't rebuild the host-list during iterating the list https://git.io/Jsvvw
09:38:00 <peter1138> It's almost like we should play this game more often...
09:39:05 <Xaroth> By the amount of bugs found, or the amount of trains crashed?
09:40:09 <peter1138> Both are rather high.
09:55:54 <DorpsGek> [OpenTTD/OpenTTD] auge8472 commented on issue #9237: I never get the new trains after 1980- but i get all plains, boats and cars https://git.io/JseBp
10:03:24 <Timberwolf> Need to figure out how to build a streaming data pipeline.
10:03:28 <Timberwolf> Coffee time, I think.
10:10:54 <LordAro> "GCC 2.9 doesn't support '::' global scope specifier when friending"
10:11:53 <TrueBrain> GCC 2.9 .. did you just say that?
10:13:28 <TrueBrain> okay, 2 hours of STUN failure research later .. and a huge tnx to Rubidium for access to a NAT that does that .. NO clue why it fails sometimes, but not always
10:13:31 <TrueBrain> like 1 out of 20 or so
10:13:39 <TrueBrain> but .. I think we just take the hit, and let those connections use TURN
10:13:47 <TrueBrain> pretty sure we can't have them all anyway
10:13:53 <LordAro> just hope that it's a low percentage
10:14:05 <TrueBrain> well, we can offload the traffic on cheap VPSes if we have to
10:14:17 <TrueBrain> I first want to have an idea how often it happens before I work on trying other ways :P
10:14:20 <LordAro> being able to get decent connection statistics out of the GC seems like a good idea
10:14:39 <TrueBrain> and I want TURN to report bandwidth stats
10:14:46 <TrueBrain> so we also know how much a game consumes
10:19:06 <peter1138> ovh-throw-away boxes...
10:19:18 <LordAro> ovh-probably-not-on-fire boxes
10:19:55 <peter1138> self-registering docker containers, let anyone can deploy an OpenTTD relay...
10:20:50 <peter1138> Hmm, shoulder still aches :p
10:21:25 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #9238: Fix: [Network] clients leaving because of broken connections was not broadcasted https://git.io/Jsv3Q
10:21:53 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #9239: Fix: [Network] don't mark the last-joined server as manual https://git.io/Jsv35
10:25:19 <peter1138> Yes, I don't know why I do that.
10:26:20 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9238: Fix: [Network] clients leaving because of broken connections was not broadcasted https://git.io/JseAw
10:26:33 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9239: Fix: [Network] don't mark the last-joined server as manual https://git.io/Jsvvn
10:27:32 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #9240: Fix: [Network] don't rebuild the host-list during iterating the list https://git.io/JsvG5
10:28:24 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9240: Fix: [Network] don't rebuild the host-list during iterating the list https://git.io/JsvZI
10:32:31 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9240: Fix: [Network] don't rebuild the host-list during iterating the list https://git.io/Jsvvw
10:36:47 <peter1138> #9241 is odd, you can place non-desert stuff in a desert, sometimes, but it doesn't convert the desert properly. I'm minded to think it shouldn't be possible regardless of normal/grove/forest option.
10:37:09 <DorpsGek> [OpenTTD/OpenTTD] vituscze opened issue #9242: Tropical Lumber Mill doesn't work properly on large maps https://git.io/Jsvnh
10:37:43 <peter1138> Not sure if the desert bit is related to rainforest status.
10:42:46 <peter1138> And OnTick_Trees() really doesn't scale by map size.
11:03:23 *** grossing has quit IRC (Remote host closed the connection)
11:05:56 <peter1138> Not having much luck with #9235 :/
11:09:09 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on issue #9242: Tropical Lumber Mill doesn't work properly on large maps https://git.io/Jsvnh
11:09:24 *** grossing has joined #openttd
11:12:40 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on issue #9242: Tropical Lumber Mill doesn't work properly on large maps https://git.io/Jsvnh
11:12:43 <FLHerne> TrueBrain: I can do more testing from here if it helps, mine also seems to do that
11:13:22 <TrueBrain> FLHerne: the issue is, it is not predictable
11:13:37 <TrueBrain> and as long as I don't understand what is causing it .. it is really hard to mitigate :)
11:13:47 <FLHerne> Try many times in succession and see what the odds are? :p
11:13:52 <TrueBrain> most of the time the first attempt fails .. after that .. pure RNG :P
11:13:57 <TrueBrain> yeah, 1 in 20, give or take
11:23:07 *** lobstarooo has joined #openttd
11:27:56 <TrueBrain> My suggestion, for "net" facility, use the classic fatal / error / warning / info for 0 / 1 / 2 / 3 log level
11:28:06 <TrueBrain> please voice your opinion if you have one on the matter :)
11:29:50 *** lobster has quit IRC (Ping timeout: 480 seconds)
11:29:59 *** lobstarooo is now known as lobster
11:30:21 *** your_opinion has joined #openttd
11:30:30 *** Rubidium sets mode: +v your_opinion
11:30:35 *** your_opinion has quit IRC ()
11:30:52 <TrueBrain> sometimes I realise the average age in this channel is not that high :P
11:33:21 <Rubidium> well add at least a 'trace'/'debug' level for things as the send frame numbers, otherwise info gets really spammy
11:33:34 <TrueBrain> send frames is debug, yes :P
11:33:39 <TrueBrain> we can add trace, sure
11:42:37 <Rubidium> I'm wondering if there's anything that's not fatal, but that should always get a message on the (in game) console
11:43:09 <peter1138> Hmm, I can fix #9242 but with a savegame bump :/
11:43:26 <Rubidium> I think (at least some) errors shown to the user in the UI already end up there
11:43:41 <TrueBrain> I think both fatal and error should go to the console tbh
11:43:48 <TrueBrain> like a "connection lost" is useful to have there too
11:43:55 <TrueBrain> but maybe that should be configurable too
11:43:59 <TrueBrain> either way .. "future work" :P
11:44:32 <Rubidium> well, you get the generic error though the lower level details might not get to you
11:46:23 <TrueBrain> Could not connect to [2001:41d0:801:2000::3297]:3975 (IPv6): Network is unreachable <- I guess we should move that error to a lower debug level too (well, higher number)
11:46:29 <TrueBrain> that is such not a problem a user should worry itself with
11:57:24 *** Elouin4 has joined #openttd
12:03:18 *** Elouin has quit IRC (Ping timeout: 480 seconds)
12:03:18 *** Elouin4 is now known as Elouin
12:05:40 <peter1138> (svn r1) Import of revision 975 of old (crashed) SVN
12:05:48 <peter1138> Okay, that's not so useful as a git blame :)
12:06:03 <LordAro> "it's been there a while"
12:11:07 <TrueBrain> so I have a dedicated game server
12:11:14 <TrueBrain> which I cannot join
12:11:24 <TrueBrain> no error nothing, it just doesn't work
12:14:13 <TrueBrain> can I get a WTH? :D
12:14:34 <Rubidium> whiskey tango hotel?
12:21:16 <TrueBrain> hahaha, okay, this is hilarious
12:21:21 <TrueBrain> I can blame Rubidium , so it really is :)
12:21:38 <TrueBrain> remember that you said: if the servers sends an invalid username, that means something bad happened to the server?
12:21:46 <TrueBrain> well .. an empty config .... sets no username for the server :P
12:22:13 <TrueBrain> so anyone joining a server where the server has set no username
12:22:23 <TrueBrain> OMG I DONT WANT TO BE HERE, response
12:22:31 <glx> maybe it should set it to "server", like "player" is used for players
12:22:34 <TrueBrain> which is pretty much correct
12:22:49 <TrueBrain> I dunno what the best course of action is tbh
12:22:54 <Rubidium> though why is the server seen as a client then?
12:23:04 <TrueBrain> a server is always also a client
12:23:09 <Rubidium> there's no-one playing ;)
12:23:11 <TrueBrain> as you can send it messages too
12:24:03 <TrueBrain> well, that took a while to figure out :P
12:25:18 *** emyaqin has joined #openttd
12:25:20 <glx> I see you found the server list "redrawing issue" I had
12:26:38 <TrueBrain> Rubidium: no worries :D It is a fun little bug .. and I found 2 bugs in my code because I was tracing this :P
12:27:21 <TrueBrain> now it is correctuped
12:28:24 <TrueBrain> for example, I found out that if there is no join-key, the game crashes .. so LAN games were impossible
12:28:38 <TrueBrain> glx: ah, yes, now you mention that .. I noticed we still have a few places that don't set the game list dirty ..
12:30:34 <glx> ok, so #9235 says there are warnings, CI says it too, but they are not inlined in the diff
12:31:21 <Rubidium> there are warnings, just not shown in the diff
12:31:26 <peter1138> Hmm, changing the fast forward speed setting does not take effect immediately.
12:31:57 <glx> github used to report them at the end if not inlinable
12:34:16 <glx> hmm and the warning is weird when I look at the reported line
12:36:13 <TrueBrain> peter1138: indeed it does not :) Was "known" at the time of writing, and considered: chances of anyone using it like that are slim
12:36:17 <TrueBrain> that might be the wrong call ofc :)
12:36:47 <glx> so annotation checks seems to be a useful thing for cases like that
12:37:32 <glx> maybe it should list the details too
12:40:56 *** grossing has quit IRC (Ping timeout: 480 seconds)
12:45:25 *** grossing has joined #openttd
12:49:03 *** sla_ro|master has joined #openttd
12:49:31 <glx> oh it's because it's not from the last commit of the PR
12:52:01 <peter1138> Hmm, do gamescripts get notified about planting trees...
12:53:26 <peter1138> Okay, maybe the performance hit was from debugging trapping and recording GS exceptions. (Every time the GS sleeps is an exception)
12:54:29 *** Feuersalamander has joined #openttd
13:07:44 *** grossing has quit IRC (Ping timeout: 480 seconds)
13:07:45 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9244: Fix: [Network] mark server as offline when no longer reachable https://git.io/Jsv7Y
13:08:18 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: multiplayer without port forwarding https://git.io/JOk7C
13:08:21 <TrueBrain> right, shitton of fixes ^^
13:08:34 <TrueBrain> just not the STUN issue some NATs show :P
13:08:36 <Xaroth> another test game tonight then? :P
13:08:43 <TrueBrain> at least TURN fallback works as expected now
13:10:24 <TrueBrain> Xaroth: only if someone prepares a nice NewGRF-game :)
13:10:38 * Xaroth points at and... wait, he's not here
13:12:36 <TrueBrain> also all sockets etc are now closed .. it should no longer leak sockets slowly :D
14:10:24 *** Flygon has quit IRC (Ping timeout: 480 seconds)
14:10:28 *** snail_UES_ has joined #openttd
14:12:15 <Speeder_> GS allow you to change how much you get paid for shipping stuff?
14:12:30 <Speeder_> also can GS affect the economy overall? to simulate crisis, booms, etc... ?
14:16:31 <glx> GS can only prevent increase/decrease of production, and closure of industries
14:27:37 <TrueBrain> meh; I do not understand why some NATs act up. All the documentation I read suggest this shouldn't be the behavior .. meh :)
14:29:19 <TrueBrain> FLHerne: what OS were you running again? I think I asked before, but I forgot :)
14:30:13 <glx> and I though I could have a weird NAT/router setup ;)
14:33:28 <FLHerne> (using NetworkManager)
14:34:06 <TrueBrain> I have been approaching this as a NAT problem .. but .. all 3 machines I know of that do this, are Linux
14:34:25 <glx> ok doing some GHA testing on my fork, I think I know why MSVC warnings are not shown
14:34:42 <TrueBrain> wait, we ruled out OS earlier, didn't we?
14:34:48 <TrueBrain> right, we did .. ugh, I should keep noted
14:34:58 <TrueBrain> it was working over your 4G
14:36:14 <FLHerne> Well, assuming it's the same issue again
14:36:44 <FLHerne> Also, I don't think "all affected clients are Linux" tells you much with the set of test players
14:37:24 <glx> there's a nice little difference in paths ;)
14:37:26 <TrueBrain> honestly, I expected more issues from our Windows players :P
14:38:42 <glx> yeah you were able to punch through 2 firewalls in my case :)
14:39:17 <TrueBrain> it is just weird .. we have an open connection from local IP:port to A
14:39:26 <TrueBrain> we set up a new connection on the same local IP:port to B
14:39:39 <TrueBrain> and somehow .. this does something ... causing the connect to report "No route to host"
14:39:46 <TrueBrain> and next attempt it is mapped differently on the NAT
14:39:54 <TrueBrain> so I guess a RST is sent for some reason
14:40:52 <TrueBrain> as from what I read, FIN shouldn't remove the mapping
14:40:59 <TrueBrain> well, not for 60 seconds at least
14:42:08 <TrueBrain> but RFC says it should even keep the mapping with an RST
14:43:10 <TrueBrain> although it says the original RFC didn't specify it, and that there are NATs out there that do reset the mapping immediately
14:43:14 *** bruvvha has quit IRC (Ping timeout: 480 seconds)
14:47:52 <TrueBrain> reading RFCs is depressing .. so many MUSTs that are not implemented :P
14:48:43 *** supermop_Home_ has joined #openttd
14:53:04 *** bruvvha has joined #openttd
14:53:23 <milek7> pebkac, but.. where do I type invite code? :P
14:55:13 <milek7> *** buffer overflow detected ***: terminated
15:02:15 <milek7> I'm trying to use invite code between two computers behind same nat, not sure if this breaks something?
15:03:26 <glx> it should work, I used client and server on the same machine
15:04:26 <glx> and client had to ask GC to get connection info
15:08:52 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: multiplayer without port forwarding https://git.io/JOk7C
15:08:56 <TrueBrain> ^^ should fix both issues
15:09:19 <TrueBrain> and your NAT doesn't implement a MUST according to RFC, and doesn't allow you to connect to yourself over your public IP
15:09:23 <TrueBrain> just so you know :P
15:10:22 <LordAro> this random game i'm running, has "Dunborough" and "Duborough"
15:10:35 <LordAro> i had to put the town windows side by side to work out what was different
15:10:51 <TrueBrain> owh, I didn't evne compile that push .. I BELIEVE IN YOU CODE
15:11:06 <TrueBrain> LordAro: well, the only question is .. why didn't you invite us to play? :P
15:11:28 <LordAro> well i was hoping to reproduce the crash
15:13:21 <FLHerne> TrueBrain: should test?
15:13:31 <TrueBrain> should test what, sorry?
15:14:23 <FLHerne> ^^ should fix both issues
15:14:33 <TrueBrain> milek7's issues, yes
15:14:49 <TrueBrain> hence the "see the line above" marker :)
15:15:09 <TrueBrain> no, if I have any clue how to solve your NAT issue, I will let you know, no worries :)
15:16:47 <TrueBrain> I made it a lot less likely for that to happen in recent PR btw; it annoyed me too much :)
15:16:58 <FLHerne> Could do with a button :p
15:16:59 <TrueBrain> and don't ask me where there is no button to delete entries ......
15:17:08 <TrueBrain> this is not new, so not part of my PR :)
15:17:20 <FLHerne> It seems to have crashed
15:19:05 <TrueBrain> I am happy you are not sure
15:19:20 *** gnu_jj_ has joined #openttd
15:19:35 <FLHerne> Window is open but not redrawing
15:19:50 <FLHerne> I clicked the window manager's (X) button
15:20:14 <TrueBrain> closing OpenTTD on Linux has signal issues
15:20:16 <FLHerne> No idea if it's related to your patches, but I haven't seen this with 1.11.x
15:20:38 <TrueBrain> so closing it via anything other than Exit Game can cause weird issues, I have noticed
15:21:02 <TrueBrain> no clue if that is what you are seeing, but it is possible
15:21:29 <FLHerne> Anyway, every thread is waiting on either ppoll() or __futex_abstimed_wait_common64()
15:21:40 <FLHerne> [whatever the latter one is]
15:21:40 <TrueBrain> sounds like a deadlock to me :D
15:22:22 <TrueBrain> looks like the audio driver is not tearing down
15:23:06 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
15:23:16 *** Wormnest has joined #openttd
15:23:47 <peter1138> Connecting to the public IP from internal IPs MUST work?
15:24:04 <peter1138> That's news to every single Linux-based NAT set up I've ever used.
15:24:26 <FLHerne> I killed it, and now my entire desktop is rendering about 2fps
15:24:50 <peter1138> Have you tried getting a new computer?
15:24:52 <TrueBrain> peter1138: I know right .. as I said, depressing
15:25:02 <TrueBrain> RFC5382, REQ-8: A NAT MUST support "hairpinning" for TCP.
15:25:47 <FLHerne> This is a relatively new computer, but my stupid problems seem to follow me :p
15:26:46 *** Progman has joined #openttd
15:27:31 *** gnu_jj_ has quit IRC (Ping timeout: 480 seconds)
15:28:37 <TrueBrain> turns out to be super easy to configure on Linux peter1138 :P
15:28:54 <TrueBrain> I also never knew btw
15:29:06 <TrueBrain> always assumed it doesn't work .. but my current NAT does allow it
15:29:10 <TrueBrain> which made testing this all a lot easier
15:31:28 <peter1138> Yeah, just the usual examples don't do it and don't mention it.
15:36:06 <Rubidium> TrueBrain: I can't find that requirement in the superseding RFC
15:37:18 <Rubidium> so maybe it's not implemented (anymore) because they implemented RFC7857 instead of RFC5382
15:38:13 <TrueBrain> This document updates RFCs 4787, 5382, and 5508.
15:39:45 <TrueBrain> So you cannot implement 7857 without 5382 ;)
15:40:48 <Rubidium> oh joy... if that's the case it sounds too much like work to me ;)
15:41:55 <TrueBrain> These RFCs were lacking details, so they kept stacking new ones to make things more clear
15:41:56 <milek7> it's not a protocol, I find it weird talking about 'implementing' nat from RFCs :P
15:42:46 <Rubidium> and I can imagine why so many NATs do it incorrectly because everybody needs to figure out which bits of a whole load of RFCs to apply or not to apply
15:43:00 <peter1138> Okay, my NAT is now RFC5382-compliant, I guess?
15:43:07 <peter1138> I wonder if it still works.
15:43:17 <TrueBrain> Which is normal for RFCs not? :D
15:43:18 <peter1138> Mine is "slightly" complex because I have some non-NAT as well.
15:45:54 <Rubidium> TrueBrain: theres also "obsoletes", see e.g. rfc2460 which obsoletes 1883 and is obsoleted by 8200
15:46:22 <TrueBrain> That is part of the fun!
15:46:44 <Rubidium> though I see many "updated by"s as well, so apparantly it became too unclear so the aggregated all the updates into 8200
15:50:26 <TrueBrain> Pretty common with RFCs honestly
15:50:38 <peter1138> Alright, now that works, now what?
15:50:45 <peter1138> I wonder what didn't work before :D
15:50:54 <peter1138> Profit? Bad feature.
15:51:12 <TrueBrain> Two SIP phones is one example :p
15:51:22 <TrueBrain> Do you have those? :D
15:51:28 <TrueBrain> You van now call each other!
15:51:33 <peter1138> No, I haven't used a non-mobile phone for years.
15:51:56 <peter1138> My landline is provided as SIP, but I never bothered connecting to it.
15:52:01 <TrueBrain> And, ofc, you can now run a OpenTTD server and client via the GC
15:52:16 <peter1138> Not least because I get free calls via mobile, and not via the landline.
15:53:35 <milek7> will SIP work behind NAT without ALG, though?
15:54:16 <peter1138> Oh right, I should remove my old employer's IPs from the management list.
15:55:12 <peter1138> Just spotted some fail2ban entries too. That's quite handy, having fail2ban block via the router instead of just the machine fail2ban is on.
15:55:47 <peter1138> I wonder if they got anywhere with IPv6 blocking.
16:00:31 <LordAro> i think that got done?
16:01:19 <peter1138> Apparently it's sunny.
16:01:23 *** gelignite has joined #openttd
16:01:38 <LordAro> quite a lot of rain and lightning here
16:02:02 <peter1138> I should, perhaps, found out if this jersey and shorts fit.
16:04:46 <peter1138> Damn, that's a spare tyre and a half :(
16:19:04 <TrueBrain> yet-another-random-NetworkDisconnect
16:19:08 <TrueBrain> our code is full of those :P
16:20:27 <TrueBrain> anyway, that one is easy to avoid: just wait a bit before you hit Join Game
16:27:18 *** frosch123 has joined #openttd
16:28:35 <TrueBrain> OverrideManagerBase issue still not fixed?
16:29:18 <TrueBrain> that bug is really annoying :P Triggers a lot of tooling to blur out a lot of warnings :D
16:29:26 <milek7> seems switching between public<>invite doesn't work
16:29:26 <milek7> it need cycling through public<>local<invite>
16:51:58 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9226: Fix #9202: Invalid test for unset NewGRF override mapping. https://git.io/JsfiB
16:52:56 <frosch123> did someone explain tb how to newgrf? :p
16:53:13 <TrueBrain> give me good GRFs to GRF with!
16:53:50 <frosch123> we should troll andy, and play with firs1 :p
16:54:00 <TrueBrain> first I need to fix another heap-use-after-free
16:57:52 <TrueBrain> so many quirks in the current network code .. I have wondered a few too many times if it isn't easier to start over :P
16:58:47 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
16:59:54 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 opened pull request #9245: Fix #9243: [Network] For a dedicated server use a fallback client and server name https://git.io/JsfXf
17:01:22 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9245: Fix #9243: [Network] For a dedicated server use a fallback client and server name https://git.io/JsfXc
17:04:55 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #9245: Fix #9243: [Network] For a dedicated server use a fallback client and server name https://git.io/Jsf1t
17:05:42 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #9245: Fix #9243: [Network] For a dedicated server use a fallback client and server name https://git.io/Jsf1n
17:09:11 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #9245: Fix #9243: [Network] For a dedicated server use a fallback client and server name https://git.io/JsfXf
17:09:51 <Rubidium> and now I can rebase my settings PR again ;)
17:10:12 <Rubidium> unless someone wants to approve it and I can merge it broken into master ;)
17:11:34 <DorpsGek> [OpenTTD/OpenTTD] frosch123 approved pull request #9226: Fix #9202: Invalid test for unset NewGRF override mapping. https://git.io/JsfMl
17:13:06 *** jottyfan has joined #openttd
17:13:10 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: multiplayer without port forwarding https://git.io/JOk7C
17:13:14 <TrueBrain> milek7: I think I fixed both issues
17:13:33 <TrueBrain> but especially the cleanup stuff is rather hard in how networking is implemented .. we will see if I had them all :P
17:17:12 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #9244: Fix: [Network] mark server as offline when no longer reachable https://git.io/JsfDi
17:18:12 *** jottyfan has quit IRC (Quit: jottyfan)
17:18:23 <TrueBrain> fucking annoying GitHub ..
17:18:50 <TrueBrain> frosch123: quick, what is the difference between {COMMA} and {NUM}? :)
17:19:15 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 updated pull request #9128: Codechange: use std::string exclusively for settings https://git.io/J3TEL
17:19:41 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9244: Fix: [Network] mark server as offline when no longer reachable https://git.io/Jsv7Y
17:21:08 <Rubidium> i.e. the thousands separator
17:21:28 <TrueBrain> that last comment of TheDude makes no sense to me
17:21:36 <TrueBrain> am I missing something?
17:23:37 <TrueBrain> because you don't assign it to a variable, it doesn't need to be resolved? Lol?
17:24:15 <Rubidium> I guess he thinks that
17:24:35 <Rubidium> though I fear in either case both of those headers are required
17:25:01 <TrueBrain> they are, I just chcked myself :P
17:27:08 <TrueBrain> right, building new binaries for 9017 .. frosch123 , found some nice NewGRFs already? :D
17:27:20 <TrueBrain> will you be hosting tonight? :P
17:27:52 <TrueBrain> takes ~20 minutes for the binaries to cook
17:28:04 <frosch123> i'll take as long to prepare a map :p
17:32:56 *** jottyfan has joined #openttd
17:42:11 <glx> let's try adding another flag to cl
17:42:33 <TrueBrain> glx: shouldn't you just look into the problem matcher and fix why it doesn't match? :D
17:42:56 <glx> just the file seems incorrect for github
17:42:57 <TrueBrain> but GitHub fails to link it?
17:43:03 <TrueBrain> lol, that is a stupid bug
17:43:26 * glx tries with the flag to output full path
17:44:50 <glx> Warning: D:\a\OpenTTD\OpenTTD\src\network\core\packet.cpp(394): warning C4267: '=': conversion from 'size_t' to 'PacketSize', possible loss of data <-- looks better, let's wait for the end of CI run
17:45:06 <TrueBrain> TheDude does ask a good question: why do we extend GameInfo to the top, not to the bottom .. that is a bit weird
17:45:07 <frosch123> TrueBrain: server is up
17:45:18 <TrueBrain> but what-ever, forward compatibility is a problem I don't want to bite into :P
17:45:22 <frosch123> someone needs to check whether all grf are on bananas, or whether i messed up :p
17:45:24 <TrueBrain> frosch123: nice :D MacOS is still ... busy
17:45:42 <glx> TrueBrain: because it's easier to fallback ?
17:45:50 <frosch123> i went for a beginner selection of newgrf :p
17:45:56 <TrueBrain> glx: yeah, the switch-statement is easier
17:47:38 <TrueBrain> it even works for your new code! :P
17:47:41 <glx> just need to do a clean patch
17:47:44 <DorpsGek> [OpenTTD/OpenTTD] embeddedt commented on pull request #9191: Codechange: Disable pointer locking by default https://git.io/Jsf7W
17:48:03 <glx> well commit checker used to work fine
17:49:04 <TrueBrain> and this is why I don't like universal binaries ........ it is like watching paint dry
17:49:46 <TrueBrain> I guess nobody should tell him we already check for -1 in many places :P :D
17:49:50 <TrueBrain> our code is so ugly sometimes :P
17:51:00 *** arikover has joined #openttd
17:54:40 <TrueBrain> (download binary, go to multiplayer, join the only server there)
17:55:34 *** andythenorth has joined #openttd
17:56:24 <andythenorth> I get the binary
17:59:00 <andythenorth> TrueBrain server is called what now?
17:59:10 <TrueBrain> not mine, and Stun something
18:00:24 <FLHerne> TrueBrain: In "Find missing newgrfs", both the missing newgrfs were "<unknown>"
18:00:36 <TrueBrain> FLHerne: takes a bit of time to fetch the data from the server
18:00:39 <FLHerne> Then I clicked "Find missing content online" and they stopped being unknown
18:00:47 <TrueBrain> I really think it should have a popup menu to show it is connecting ..
18:00:51 <TrueBrain> but .. yeah .. one thing at the time :D
18:09:43 <peter1138> Hmm, is there something like an UnscaleByMapSize, like we have for zoom levels? :D
18:13:44 <DorpsGek> [OpenTTD/OpenTTD] unbeatable-101 commented on issue #9236: Vehicles do not pass stopped/broken down vehicles https://git.io/J3jij
18:42:13 *** iSoSyS has quit IRC (Remote host closed the connection)
18:49:43 *** jottyfan has joined #openttd
18:56:23 <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #9246: Fix #9242: Tree tick handler did not scale by map size. https://git.io/Jsfjz
18:57:15 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #9246: Fix #9242: Tree tick handler did not scale by map size. https://git.io/Jsfj6
19:01:05 *** lobstarooo has joined #openttd
19:03:19 <DorpsGek> - Update: Translations from eints (by translators)
19:03:49 <frosch123> the eints time still surprises me :) what, so early?
19:06:36 *** lobster has quit IRC (Ping timeout: 480 seconds)
19:06:36 *** lobstarooo is now known as lobster
19:07:19 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #9246: Fix #9242: Tree tick handler did not scale by map size. https://git.io/JsJep
19:16:18 <peter1138> Oh, my local regression test no longer works :/
19:19:52 <peter1138> Hmm, becuase the regression stuff is not installed?
19:30:11 <peter1138> Okay so regression is 256x256 so it'll just be because the tick starts earlier. Hmm.
19:32:25 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #9247: Fix: [Actions] Annotations not shown for MSVC https://git.io/JsJUA
19:38:36 <TrueBrain> frosch123: I like these NewGRFs, it makes the game more interesting
19:38:44 <TrueBrain> you need to get farming supplies to boost production
19:38:48 <TrueBrain> but that needs engineering supplies
19:38:51 <frosch123> glad i did not overload you :)
19:39:12 <TrueBrain> its complicated, but you just have to start somewhere
19:39:27 <frosch123> next time yeti+nuts :p
19:39:45 <TrueBrain> something that does something more interesting with towns would be nice
19:39:47 <frosch123> TrueBrain: i think, fanboys of "firs extreme" just like rediscovering routes everytime :)
19:40:18 <TrueBrain> one thing I absolutely hate, is forgetting to refit
19:40:34 <frosch123> yep, that's a trap :)
19:40:42 <andythenorth> TrueBrain get grfs that support station refit
19:40:45 <frosch123> but you can use the vehicle list fitler to select a cargo
19:40:48 <peter1138> If you filter by cargo type it'll refit
19:40:52 <frosch123> that shows you the wagons, and also refits on build
19:40:55 <andythenorth> refit in the refit UI is stupid yak-shaving
19:41:01 <andythenorth> it's a horrible UI
19:41:07 <TrueBrain> yeah, I am now doing station refit
19:41:11 <TrueBrain> which feels cheating or something
19:41:13 <peter1138> Which was a lovely desync fest when I first did it... all wrong.
19:41:16 <TrueBrain> also, it is not really clear if it costs you anything
19:41:21 <andythenorth> but a lot of people think it is
19:41:37 <TrueBrain> I guess it is free?
19:41:41 <andythenorth> like...it's cheating because you don't have to click through some awful UI
19:41:57 <TrueBrain> it feels like cheating as that is not how the world really works
19:42:03 <andythenorth> refitting once in depot is so weird
19:42:04 <TrueBrain> you don't refit a wagon from oil to water n a station
19:42:06 <FLHerne> You mean the orders UI isn't awful?
19:42:30 <TrueBrain> but I really like the Enhanced and Gung-ho
19:42:33 <TrueBrain> that is just an awesome concept
19:42:40 <peter1138> auto-refit regearing?
19:42:50 <TrueBrain> and I love that airports are expensive as fuck
19:44:22 *** jottyfan has quit IRC (Quit: jottyfan)
19:44:23 <andythenorth> peter1138 now that would be silly
19:44:31 <andythenorth> also....can't auto-refit subtypes
19:44:59 <andythenorth> oh that's why people think it's cheating
19:45:07 <andythenorth> they can't set the correct livery with station refit
19:46:40 <_dp_> does firs have a storage for supplies now?
19:47:13 <TrueBrain> Xaroth has a nice idea, when someone else is joining the map, also show the other players (when paused) how far the transfer is
19:47:22 <_dp_> what's that 0 waiting thing then?
19:47:23 <TrueBrain> the data is there , just a matter of GUI and some packet to inform everyone
19:48:07 <frosch123> _dp_: we are playing old school firs
19:48:30 <frosch123> back then it always showed "0 waiting", for any accepted cargo
19:48:41 <TrueBrain> also fun fact, 50% of the current players have IPv6 :P
20:00:23 <TrueBrain> but I dare say, that this PR is nearing completion ..
20:00:31 <TrueBrain> just the Game Coordinator ... but that can wait :P
20:00:38 <TrueBrain> the current version works sufficient for nightlies
20:04:35 <LordAro> is the GC code anywhere public yet? :p
20:04:47 <TrueBrain> no, but if you like, I can make that happen
20:17:19 <peter1138> Regression test regression does not exist (tip: check regression folder for the correct spelling)
20:18:29 <peter1138> Where does it all need to live? :(
20:24:10 <DorpsGek> [OpenTTD/OpenTTD] embeddedt commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/JsJ3v
20:26:40 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/JsJ3R
20:27:45 *** Wormnest has quit IRC (Quit: Leaving)
20:28:30 *** EER has quit IRC (Ping timeout: 480 seconds)
20:28:42 <peter1138> Okay, I need to copy regression/* to build/ai/
20:38:12 *** orudge has quit IRC (Remote host closed the connection)
20:40:43 <DorpsGek> [OpenTTD/OpenTTD] embeddedt commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/JsJGz
21:02:11 <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #9246: Fix #9242: Tree tick handler did not scale by map size. https://git.io/Jsfjz
21:07:34 <peter1138> Hmm, is it okay that scenario editor tools bypass the command system?
21:07:53 <frosch123> no multiplayer-scenario editor :)
21:08:22 <frosch123> but yes, cheats and SE bypass the commands
21:11:27 <glx> <@peter1138> Okay, I need to copy regression/* to build/ai/ <-- but cmake should copy everything where it needs to be
21:11:53 <peter1138> glx, is there a step I'm missing to do that?
21:12:19 <glx> it's done during build I think
21:12:26 *** gelignite has quit IRC (Remote host closed the connection)
21:12:39 *** gelignite has joined #openttd
21:14:02 *** andythenorth has quit IRC (Quit: andythenorth)
21:14:09 <peter1138> Hmm, PlaceTreeGroupAroundTile() doesn't appear to "be a bit picky about which trees go where"
21:15:03 <peter1138> Nor call SetTropicZone() if appropriate.
21:19:09 <DorpsGek> [OpenTTD/OpenTTD] embeddedt opened pull request #9248: Fix: [Emscripten] Force secure WebSockets over HTTPS https://git.io/JsJWz
21:21:34 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on pull request #9248: Fix: [Emscripten] Force secure WebSockets over HTTPS https://git.io/JsJWS
21:22:20 *** Samu has quit IRC (Quit: Leaving)
21:22:56 *** gelignite has quit IRC (Quit: Stay safe!)
21:24:42 <DorpsGek> [OpenTTD/OpenTTD] embeddedt commented on pull request #9248: Fix: [Emscripten] Force secure WebSockets over HTTPS https://git.io/JsJ8v
21:26:39 *** frosch123 has quit IRC (Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn)
21:30:15 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
21:34:34 *** SpComb has quit IRC (Server closed connection)
21:40:16 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:46:40 *** berndj has quit IRC (Server closed connection)
21:47:45 *** SpComb has quit IRC (Read error: Connection reset by peer)
21:47:52 *** Tulitoma1tti has joined #openttd
21:48:30 *** Tulitomaatti has quit IRC (Ping timeout: 480 seconds)
21:57:21 <LordAro> i've been semi-playing a game for most of the evening (it's spent most of the time on FFWD), but i have finally managed to make a non-city have the largest population (despite still serving all other towns)
21:59:06 <_dp_> it's kind of how every citybuilder game goes :p
21:59:18 <_dp_> as only towns are claimable
22:07:56 <LordAro> oh this is entirely stock
22:08:11 <LordAro> (well, technically it's TB's STUN branch, but also single player)
22:12:16 <LordAro> my trains are a bit of a mess, but it's a mountainous alpine map
22:12:39 <_dp_> stock just means there are no extra requirements for growing towns :p
22:12:57 <_dp_> and you can just powerfund it to whatever
22:13:37 *** Hazzard has quit IRC (Server closed connection)
22:13:46 <LordAro> i'm not exactly trying
22:13:58 <LordAro> have spent most of the evening on a zoom call and drinking
22:13:58 *** XeryusTC has quit IRC (Server closed connection)
22:14:04 *** Xaroth has quit IRC (Quit: Ping timeout (120 seconds))
22:14:19 *** Webster has quit IRC (Server closed connection)
22:14:38 *** Hazzard has joined #openttd
22:15:07 *** XeryusTC has joined #openttd
22:15:16 *** EER has quit IRC (Ping timeout: 480 seconds)
22:15:20 *** Webster has joined #openttd
22:20:10 *** sla_ro|master has quit IRC ()
22:32:02 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
22:33:20 *** Tirili has quit IRC (Remote host closed the connection)
22:36:33 <LordAro> did the news ticker not used to scroll really fast on FFWD?
22:36:38 <LordAro> or am i making that up?
22:38:49 *** Markk has quit IRC (Server closed connection)
22:40:24 *** WormnestAndroid has quit IRC (Remote host closed the connection)
22:41:46 *** WormnestAndroid has joined #openttd
22:47:14 *** Progman has quit IRC (Remote host closed the connection)
22:47:49 *** lastmikoi has quit IRC (Server closed connection)
22:48:51 *** Markk_ is now known as Markk
22:51:23 *** lastmikoi has joined #openttd
22:53:49 *** Timberwolf has quit IRC (Server closed connection)
22:58:52 *** Timberwolf has joined #openttd
23:00:09 *** Heiki has quit IRC (Remote host closed the connection)
23:02:27 *** arikover has quit IRC (Quit: ERC (IRC client for Emacs 27.1))
23:07:36 *** michi_cc_ has joined #openttd
23:08:34 *** michi_cc has quit IRC (Remote host closed the connection)
23:18:19 *** avdg has quit IRC (Server closed connection)
23:18:49 *** Osai has quit IRC (Server closed connection)
23:19:19 *** ^Spike^ has quit IRC (Server closed connection)
23:19:47 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:20:21 *** ^Spike^ has joined #openttd
23:21:49 *** moll has quit IRC (Server closed connection)
23:28:54 *** WormnestAndroid has joined #openttd
23:41:04 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:43:20 *** WormnestAndroid has joined #openttd
23:53:18 *** Alkel_U3 has quit IRC (Quit: maintenance)
23:56:13 *** Alkel_U3 has joined #openttd
continue to next day ⏵
