IRC logs for #openttd on OFTC at 2022-08-26
⏴ go to previous day
00:43:27 *** giords[m] has joined #openttd
00:47:50 *** cjmonagle[m] has joined #openttd
01:12:57 *** cawal[m] has joined #openttd
01:42:30 *** karl[m]123456 has joined #openttd
02:18:53 *** Wormnest has quit IRC (Quit: Leaving)
02:56:46 *** debdog has quit IRC (Ping timeout: 480 seconds)
02:59:44 *** olmvnec[m] has joined #openttd
04:40:20 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
04:40:23 *** WormnestAndroid has joined #openttd
04:50:43 *** smoke_fumus has joined #openttd
04:56:01 *** D-HUND is now known as debdog
05:14:38 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
05:17:02 *** WormnestAndroid has joined #openttd
06:23:42 *** philip[m]1 has joined #openttd
06:24:58 *** kolcon1 has quit IRC (Ping timeout: 480 seconds)
06:34:53 *** kolcon1 has joined #openttd
06:44:02 <LordAro> but why does it crash?
06:44:16 <LordAro> size isn't underflowing, surely?
06:44:56 <LordAro> oh wait, stack overflow
06:46:52 <nielsm> I'm thinking it should be replaced with a std::deque
06:47:15 <nielsm> and maybe truncating to something reasonable once in a while
06:47:40 <LordAro> does it need a deque? a vector would suffice, surely?
06:48:30 <LordAro> or, hmm, cutting elements off the front of a vector isn't necessarily particularly fast, is it?
06:48:40 <LordAro> it is if it's just an array :p
06:48:56 <nielsm> yeah cutting off the front of a vector requires moving everything up
06:49:12 <nielsm> a deque is designed to be efficient for adding and removing on both ends
07:11:56 <nielsm> IConsolePrint(TC_LIGHT_BLUE, "OpenTTD Game Console Revision 7 - {}", _openttd_revision);
07:12:17 <nielsm> does changing the internals of the console warrant bumping that revision 7 to 8, perhaps?
07:12:25 <nielsm> the external interface is unchanged
07:25:46 <dwfreed> if the behavior is nominally the same, I would say no
08:03:32 *** dP has quit IRC (Remote host closed the connection)
08:03:43 *** DorpsGek_v has joined #openttd
08:04:11 *** TrueBrain has joined #openttd
08:04:11 <TrueBrain> funny, it seems all puppets were connected to the same IRC host 😄 Well, guess that happens with GeoDNS or something 😉
08:04:16 <TrueBrain> either way, we are back jfs- 🙂
08:04:38 <LordAro> nielsm: some warnings for you to fix
08:04:44 <nielsm> the discord bridge died
08:04:56 <TrueBrain> no, IRC had a netsplit! 😛
08:07:20 <peter1138> Hmm, only 50 miles planned tomorrow morning, that means it'll be... pacey.
08:07:54 <LordAro> trouble is you've proved you can do it now
08:11:38 <peter1138> I'll bring a rope, literally towed :D
08:12:11 *** virtualrandomnumber has joined #openttd
08:12:30 *** virtualrandomnumber has quit IRC ()
08:12:46 <peter1138> Going past Silverstone, that's going to be busy.
08:15:26 <dwfreed> TrueBrain: make sure the bridge is resolving irc.oftc.net for every connection
08:15:36 <dwfreed> TrueBrain: and not just resolving it once and reusing that
08:16:05 <TrueBrain> I did exactly that 🙂
08:16:19 <TrueBrain> the OS can still cache ofc
08:16:33 <dwfreed> glibc does no caching
08:16:49 <dwfreed> unless you're running nscd, which would be the actual cacher there
08:16:51 <TrueBrain> no, but nss could .. but I don't expect that to be active in the AWS images 🙂
08:17:58 <TrueBrain> but despite that, it looks all but one connection was made to the same IRC host 🙂
08:22:58 <TrueBrain> more the question is I guess why it took 10-ish hours before the first puppet noticed it was no longer connected to IRC .. I need to look into this part of the irc-library I am using
08:24:57 <nielsm> the irc protocol has a PING command clients can send, and the server responds back with whatever parameter the client gave
08:25:05 <nielsm> no idea if that's any use here
08:32:15 <nielsm> wow, the msys2 build checks take ages to run
08:33:13 <nielsm> also is it time to put an ARM based build in the pipeline?
08:33:51 <nielsm> actually, is the mac build already dual-arch?
08:38:26 *** NGC3982_ is now known as NGC3982
08:39:42 <nielsm> right, on the topic of irc ping, I'm pretty sure that most ircds already do send a PING to the client if there hasn't been any other communication with it for a while, and closes the connection if there's no response
08:40:30 <dwfreed> yeah, the ircd sends pings to idle clients
08:40:46 <dwfreed> but clients can also send pings to make sure the connection is still alive
08:42:48 <dwfreed> idle TCP sockets produce no traffic (even with SSL), and the particular failure mode here was that all network connectivity to the ircd was just completely dropped, so no resets made it through, and thus clients would not know the connection was dead unless they were checking
08:42:51 <peter1138> If the client knows the server will send a ping, then it can also determine its connection is lost by the lack of ping.
08:43:01 <peter1138> But that's dependent on the client library.
08:43:20 <dwfreed> peter1138: this would need to be heuristically determined, as different networks have different ping times
08:43:42 <dwfreed> and for a rather active client, you may never actually see a ping from the ircd
08:44:15 <dwfreed> (libera's ping time is 120 seconds since the last activity from the client, oftc's is 240 seconds)
08:44:52 <dwfreed> both libera and oftc only send pings if the client hasn't sent them anything in that timeframe
08:49:50 <nielsm> nearly 40 minutes for the msys2 compile check
09:25:03 <Hutchinson[m]> Hello, Are you interested in making $1,500 plus additional $500 for diligence and hardwork in two weeks (legit) by sparing just 15/30 minutes of your time every 48hrs without no start up fee ? If yes get back to me for more details
09:25:15 *** peter1138 sets mode: +b *!*hutchins_@2001:470:1af1:101::*
09:25:17 *** Hutchinson[m] was kicked by peter1138 (Hutchinson[m])
09:26:30 <dwfreed> peter1138: murdered, you can remove the ban (it's a bit too generic, anyway)
09:27:25 <peter1138> It's a single /64 block, and has a username in it.
09:28:01 <dwfreed> and covers every matrix user who's name begins with hutchins_
09:28:08 <dwfreed> s/who's/whose/ words are hard
09:29:05 <dwfreed> that single /64 is everybody using the matrix.org bridge
09:29:09 <peter1138> That's their problem.
09:29:39 <dwfreed> the spammers are never the same twice, so there's no point in keeping it
09:30:24 <peter1138> This is more spammy than the spammer. The ban is on a timer, it'll go.
09:31:13 *** peter1138 sets mode: -b *!*hutchins_@2001:470:1af1:101::*
10:08:38 *** WoozyDragon4018 has joined #openttd
10:52:42 <LordAro> oh, lol, CI disagrees
10:55:40 <nielsm> gah... I'll look at it in a bit, working on something else rn
10:56:17 <LordAro> i'd almost suggest that the whole global debuglevel array should be converted to a map
10:56:21 <LordAro> but that's probably a bit much
10:56:36 <LordAro> would save the .find inside a loop though
10:57:11 <LordAro> could just do debug_level = new_levels :)
10:57:25 <nielsm> nah doesn't work like that
10:57:40 <nielsm> the actual debug levels are stored in individual global int variables
10:57:54 <nielsm> the DebugLevel struct has a _pointer to_ the debug level value
10:58:04 <LordAro> but they don't have to be?
11:07:28 <peter1138> Oh shit it's lunch time already. And I just had a snack, oops.
11:43:28 <nielsm> LordAro: the const issue is fixed now!
11:49:27 <nielsm> I think that's enough code work for today
11:51:41 <peter1138> I had "truffle flavoured" smoked salmon. Such posh.
11:55:52 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:56:51 *** WormnestAndroid has joined #openttd
12:02:15 *** ely-the-kitsune has quit IRC (Ping timeout: 480 seconds)
12:06:02 *** andythenorth has joined #openttd
12:06:02 <andythenorth> FLHerne: Does it have flat docks?
12:06:34 <FLHerne> the dock is itself a boat, so I guess so :p
12:06:39 <andythenorth> Are these on the Trent at Attenborough?
12:07:47 <FLHerne> No, on the Severn at Upton-on-Severn
12:08:19 <FLHerne> it's an old barge repurposed as a big mooring pontoon, with a conveyor over top
12:08:27 <FLHerne> I haven't passed the unloading point yet
12:08:40 <andythenorth> These are sort of in FISH ish
12:08:57 <andythenorth> I used them for some tiny barge thing
12:09:22 <FLHerne> they look a bit bigger to me than FIRS' really tiny barge
12:09:50 <FLHerne> scale is weird though
12:10:20 <FLHerne> tbh I've never had a good idea of what size of vessels FISH is supposed to be, I just buy the ones that look right :p
12:11:08 <FLHerne> hm, probably it is the size of the FISH tiny-barge, just different proportions
12:11:14 <FLHerne> these ones are short and wide
12:11:33 <FLHerne> roughly 200t load I think
12:11:48 <FLHerne> the internet said 350 but that seems clearly wrong
12:12:07 <FLHerne> even modelling it as a cuboid across max dimensions it doesn't displace that much water
12:12:48 <FLHerne> here comes another one now
12:13:19 <andythenorth> Boats scale is hard to judge
12:13:28 <andythenorth> Bulldozers are the same
12:14:09 <andythenorth> From a photo with no context a ship could be 1000 grt or 10000 grt
12:14:44 <FLHerne> yeah, but in this case I'm viewing them repeatedly from a few yards away around several other boats of known dimensions ;-)
12:15:17 <andythenorth> I think the FISH one was 40-80 ft
12:16:09 <FLHerne> they're about 70ft long and 20ft wide
12:16:23 <FLHerne> and 4-5 feet of the side disappears when loaded
12:30:02 *** ely-the-kitsune has joined #openttd
12:50:45 <peter1138> That's a long fish. How much did it weigh?
12:58:06 *** nielsm has quit IRC (Ping timeout: 480 seconds)
13:45:47 *** NickSpielt has joined #openttd
13:48:41 *** CleoWaterdev has joined #openttd
13:49:02 <CleoWaterdev> Discord Moderator: sry for ping but pls remove
13:50:09 *** Chris[d] has joined #openttd
13:50:09 <Chris[d]> I've seen less convincing scams
13:50:10 *** Brickblock1 has joined #openttd
13:50:28 <CleoWaterdev> Chris[d]: yea but still haha
13:51:19 <peter1138> "got hacked" got tricked into giving permissions
13:53:42 <CleoWaterdev> peter1138: got tricked to clicking on a link and entering discord credentials lmaoo
15:30:32 *** HerzogDeXtEr has joined #openttd
15:44:11 *** virtualrandomnumber has joined #openttd
15:44:26 *** virtualrandomnumber has quit IRC ()
16:29:59 *** gelignite has joined #openttd
16:45:10 *** Wormnest has joined #openttd
17:31:50 *** Wormnest has quit IRC (Quit: Leaving)
17:49:16 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
17:54:37 *** WormnestAndroid has joined #openttd
18:32:23 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
18:58:34 <DorpsGek> - Update: Translations from eints (by translators)
18:58:56 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
18:59:09 *** WormnestAndroid has joined #openttd
19:07:46 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
19:10:08 *** WormnestAndroid has joined #openttd
19:50:50 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
19:52:03 *** WormnestAndroid has joined #openttd
19:58:33 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
21:19:43 *** gelignite has quit IRC (Quit: Stay safe!)
22:30:30 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
22:35:33 *** WormnestAndroid has joined #openttd
22:55:36 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
continue to next day ⏵
