Time Nick Message 00:01 Grandolf what are the legalities of builds in a world? am i allowed to run someone elses server if im given the files? and is it legal for the files to be transferred? 00:03 Grandolf also is it legal for me to capture someones builds in a server? 00:06 sfan5 I don't think there's any lawyer here than can answer these questions 00:06 Shara Grandolf: Will anyone take you to court over copying builds? I doubt it., but also not a lawyer. However some people to set a license or terms of use on schematics they make available, and some players will certainly be very upset if you go and copy their work without their agreement. 00:07 Shara do set* 00:08 T4im abstractly speaking any work that passes a threshold of originality could potentially be copyright protected; whether that is expressed on paper or a minetest server ought to make no difference 00:09 T4im but i wonder if in the minecraft world that actually has ever been taken to court :x if not, then i highly doubt it's gonna happen here :P 00:10 Shara I do have something in my server rules saying original builds should be considered copyright of whoever made them and you should ask before reusing them for something outside my servers. But some things are about decency rather than whether they can be enforced or not. 00:11 T4im indeed 00:11 T4im i mean, asking for permission is bothe decent and removes any legal doubts one might have 00:11 T4im :D 00:16 Grandolf its difficult to ask every memeber who ever built in a world if its ok to host an old server again 00:17 Grandolf especially when that server has had alot of people in it who dont play minetest anymore 00:18 T4im if you have the permission of the previous hoster, it's usually ok; as long as you don't create a copy of the world parallel to the original one; which hasn't worked out well in the past 00:19 Grandolf if someone hosted the server, and someone else did all the modding and admin-ing, whos is the server? 00:19 Grandolf who owns it? 00:21 T4im that seems irrelevant; if your goal is to keep the world alive, you ideally want the permission of anyone involved in hosting/administration to agree you taking over 00:21 T4im so that noone in the previous team gets the idea to do the same themselves, and you end up with two versions of the world online at the same time 00:21 T4im because that really messes things up 00:24 T4im this is much more about being pragmatic than legal questions in my opinion :D 00:24 Grandolf would you belive me if i saw this exact situation DX 00:25 Grandolf except one group wants to run it and the other wants it to stay down 00:27 T4im well, that doesn't sound good 00:29 Grandolf they use to be good friends too :P 00:29 Grandolf highlight the word use 00:30 BillyS Japanese message don't show in chat due to their wierd characters 00:30 BillyS Is this easily fixed? 00:31 BillyS I would assume not ... it would probably require all the clients to use a different font 00:44 Shara Grandolf: Sometimes when you have a dispute like that, it's better not to get involved at all. 09:12 VanessaE sofar: please close https://github.com/minetest-mods/biome_lib/issues/4 09:13 VanessaE (invalid issue, no replies for over a year, video link is dead) 10:20 VanessaE (not to mention, it's on the wrong fork) 11:35 Krock hi tenplus1 who isn't here again 11:35 T4im hi krock 11:35 T4im :) 11:35 Krock hi T4im :D 11:39 xerox123_ !seen tenplus1 11:39 MinetestBot xerox123_: tenplus1 was last seen at 2018-11-24 15:38:18 UTC on #minetest-hub 15:00 IhrFussel If you release 0.4.17.2 or 0.4.18 then PLEASE fix the map stalling 15:01 rubenwardy when did the problem start? 15:01 IhrFussel Someething in the MT code allows certain clients (AFAIK only mobile clients with random app names) to make the map STOP loading for everyone, but mods still function as expected 15:03 IhrFussel Pretty sure it was a few months ago... it's either a rather popular badly programmed app or some kind of bots... but whenever a certain user with a name like "[A-Z][a-z]{1,}[0-9]{3,4}" disconnects everything goes back to normal ... there must be some security hole in the map sending code that allows such a blocking 15:04 IhrFussel Inventories are affected too then it seems...so map + meta maybe 15:05 Krock something like slow loris attack? 15:07 VanessaE you know, I had that problem for a while, too 15:07 VanessaE eventually it just stopped happening though 15:07 VanessaE no idea (at the time) why it happened, or what made it stop. 15:08 IhrFussel VanessaE, it's only some clients that cause it and it only happens on very popular servers mostly... maybe it even takes multiple of these "badly programmed clients" until the stalling occurs... first it slows down map loading and later it stops completely 15:08 rubenwardy map sometimes stops responding on CTF 15:09 rubenwardy I wonder if it's something to do with reliable sending or using too much resources on one client? 15:09 IhrFussel I saw it happening there too...maybe some with that bad app firgured out how to write their own names and connected to it...not sure 15:10 VanessaE https://github.com/minetest/minetest/issues/6009 15:10 IhrFussel AFAIK these apps (that give a random name) let you type your own but 99% of the users don't know or don't bother to change it 15:11 IhrFussel I can see if I still have the videos I made for proof 15:12 Shara I've been having issues with map loading recently as well since activity stepped up a bit on RC... and I've noticed it seems to conincide with certain players joining and leaving as well. 15:12 Shara I've done checks while it's happening to make sure it's not likely to be people exploring or heavy mesecons use. 15:13 T4im hm, might be a bug in their forks that causes this; on BuckarooBanzai's server a lot of those mobile users just kept running into the same damn direction every time; jumping of a platform, pretty much all landing in similar areas far off 15:14 T4im was way too precise to not be a client bug there :p 15:15 * T4im makes the connection between erranous movement and the slowdown due to mapblock sending requests 15:15 Shara Random hunch based on the player that I think caused it being able to see my name tag even when I hid it - possibly an app using a pretty old version. 15:16 rubenwardy nice pixels there 15:16 Krock that would be 0.4.11 or earlier 15:16 ANAND That's possible on 0.4.13 too 15:16 Shara Krock: No, 4.13 I think 15:16 Shara Grr, ANAND is quick :) 15:16 Krock quick and correct for the 0. prefix 15:16 ANAND :P 15:23 IhrFussel Is it possible to "cut" the sending of mapblocks if it takes too long for the client to respond? 15:27 T4im well, you could start by checking whether those lags are mapgen caused 15:28 T4im as in, the clients causing mapgen 15:30 Krock hm.. there's only one send thread? 15:31 rubenwardy #7895 15:31 ShadowBot https://github.com/minetest/minetest/issues/7895 -- [WIP] Add luacheck to check builtin by rubenwardy 15:34 T4im why exclude the mainmenu though? :p i know it's exploding with warnings; but since this isn't in travis, it'd give people the opportunity to fix those :D 15:34 T4im without failing builds 15:34 T4im i bet, if you exclude it, it'll never be fixed :x 15:34 rubenwardy I'm planning on adding it to travis in that PR 15:34 rubenwardy heh 15:34 rubenwardy maybe 15:35 rubenwardy the worst in c55's code is local i for i=1, #thing do 15:35 T4im :D 15:35 rubenwardy sapier's code, ie: mainmenu, is like 100s of warnings 15:36 rubenwardy tbh, most of them are non issues really 15:36 T4im lemme guess, the profiler, besides its defined top-level global has none? :x 15:36 rubenwardy there was a single line being too long 15:36 T4im :o 15:37 rubenwardy but apart from that, no issues 15:39 rubenwardy already halved the number of mainmenu issues just be saving some files to remove whitespace, and adding table things to config 15:43 T4im hmm you might want to consider do-end blocks instead of reassigning variables for clearer scoping 15:43 T4im the setname/setvalue thing 15:44 T4im won't make a big difference here though, except for communicating intent 15:47 * T4im didn't do that when adding linting to mtg :/ 15:51 rubenwardy setting non-standard global variable _ 15:51 rubenwardy u wot 15:53 T4im in item.lua it seems 15:53 T4im you removed the local _ in front of the for loop 15:55 rubenwardy this is somewhere else 15:56 IhrFussel T4im, it#s not mapgen, all of the above posted videos were recorded in an already generated and well-known area 15:57 T4im IhrFussel: you were in that already generated area; but were the guests? 16:02 IhrFussel But it's map sending then not generating...clients don't generate map, they only receive it 16:02 T4im but they can trigger it 16:02 IhrFussel But if one client generated it in the past then there is nothing to generate anymore 16:03 T4im theres always someone new to generate 16:03 IhrFussel Mapgen only triggers once per map lifetime unless you delete mapblocks from the DB 16:03 T4im something* 16:03 IhrFussel per mapblock* 16:03 T4im per chunk i believe, but yes 16:04 T4im (i.e. 5x5x5 mapblocks iirc) 16:04 IhrFussel And these areas are 2 years old so not new ... the problem is clients receiving the map...and somehow managing to make map loading stop for everyone 16:04 T4im the idea here is, that if someone had the clever idea to spawn mobile users into random places in the world, they'd essentially DoS your server 16:05 T4im or if somehow their movement code is broken and the player teleports into random places, that could happen, too 16:05 IhrFussel Then NoCheat should kick in AFAIK...let me check 16:08 T4im no idea what work has been done there in the last years, but as long as the client is in charge of physics, i'd keep worrying about it 16:08 IhrFussel There are lots of "moved too fast; resetting position" lines but not sure if it's related...it also happens with slow connections 16:08 T4im does it mention the positions? 16:09 IhrFussel Nope but it can happen quite a few times per second 16:14 IhrFussel ~ 80,000 lines of that since nov 16 16:14 T4im well; certainly enough to check if mapgen starts up the next time you have that issue 16:14 T4im because mapgen *does* have that effect 16:15 T4im so at the very least excluding the possibility would be worth it 16:17 IhrFussel I guess I can add a register_on_mapgen() log entry 16:17 T4im :) 16:23 IhrFussel Does MT really allow mapgen to take place at any position even if the player shouldn't be there yet? 16:23 rubenwardy no 16:24 Krock kinda, since it generates a defined radius around the player 16:25 IhrFussel Yes but MT should check if the player could've been at that position first 16:25 Krock yes, that's what anticheat does 16:25 IhrFussel And it should abort mapgen if it#s false 16:27 IhrFussel So I guess the last part isn't implemented 16:27 Krock hi tenplus1 16:27 tenplus1 hi folks, and Krock 16:27 Krock HAH 16:27 tenplus1 snap! 16:27 T4im hi :) 16:27 tenplus1 ehehe, hi T4im 16:27 Krock this time. this single time I did it 16:27 tenplus1 :P 16:29 T4im the question might very well be if mapgen would be started before or after such a correction; could be an unintented effect 16:29 tenplus1 ? 16:29 T4im to the conversation before you joined :x 16:30 T4im abut mobile users bringing mapsending to a halt 16:30 tenplus1 ahh, it does lag more with mobile users, yeah 16:30 IhrFussel WAIT 16:30 tenplus1 hi fussel 16:30 IhrFussel But mapgen halts mod code correct? AFAIK it lags the whole server 16:31 rubenwardy no 16:31 rubenwardy mapgen is on a separate thread 16:31 rubenwardy it halts mod code when register_on_** is called 16:31 tenplus1 hi ruben... register_on is used in quite a few mods tho 16:31 T4im on_mapgen 16:31 rubenwardy it's only as slow as the code inside the mapgen callback 16:32 IhrFussel But on my old server machine (Opteron) I had a 1-2 sec lag whenever I generated new map parts... not sure if that could've been caused by a mod 16:33 IhrFussel And when mapgen runs mapsend cannot run?? 16:34 IhrFussel The problem is clearly map sending... so it only makes sense if mapgen really blocks any sending of map 16:34 IhrFussel Hi tenplus1 16:34 tenplus1 o/ 16:35 T4im from my experience it does 16:36 IhrFussel So it mainly only happens with appname-users, it only happens when many are online and it happens in "steps" (first map gets slower and slower until it halts) ... sounds like some sort of looping to me at least 16:36 T4im whenever someone is exploring you may end up running against CONTENT_IGNORE nodes 16:36 T4im or whatever the client is using in that place 16:37 T4im well; the mapgen idea is just one thing to rule out; could just as well be, that the anticheat itself is the cause 16:37 tenplus1 I disable anticheat on server 16:37 IhrFussel AntiCheat happens in the thread where mod code is run no? But mod code runs perfectly fine 16:37 T4im maybe the client teleports, anticheat teleports back; client teleports again, and again, and again :p 16:37 tenplus1 and rollback 16:39 IhrFussel The chat, entities, ingame time/events...they all work and max_lag also doesn't increase 18:06 tenplus1 laters folks 19:36 IhrFussel Happened again...so yeah it's definitely ONE specific player who stalls the map... and it doesn't matter if other join or leave only when THAT one user leaves the game loads the map again 19:37 IhrFussel I couldn't restart yet, but will do so a bit later...then I'll see if the stalling occurs during lots of register_on_generated() callbacks 22:02 IhrFussel Is player speed calculated server side? Like pos1 -> pos2 = movement speed? 22:02 IhrFussel I hope MT doesn't trust the client there 23:01 GreenDimond I have a registered test LBM and it is doing nothing 23:02 GreenDimond Wait 23:02 GreenDimond im dumb