Time Nick Message 02:56 Blockhead256[m] ipv6 is a real issue and I'm not blaming the Minetest devs necessarily, but at one point I had to factory reset my home router to un-bork its IPv6 and be able to connect to my server 02:57 Guest54 > not blaming the Minetest devs 02:57 Guest54 ha, we all know minetest devs are at fault for developing ipv6 in the first place! 02:58 Blockhead256[m] well my ISP deserves no sympathy for crappy router either 08:04 appguru celeron55: I'd like to become a helper on the Minetest Forums (I'm LMD: https://forum.minetest.net/memberlist.php?mode=viewprofile&un=LMD). There are now 4 pages of topic move requests for mods. LRV / Mooncarguy who was originally made a helper (to alleviate this IIRC?) seems to have since become inactive (see https://forum.minetest.net/memberlist.php?mode=viewprofile&u=20302). 08:33 sfan5 @Warr1024 ipv6_server = true will listen on both AFs 08:34 sfan5 but if you set that you need to make sure v6 does work or the serverlist rejects you 08:34 sfan5 Blockhead256[m]: you can blame us for not implementing ipv6 happy eyeballs or any fallback mechanism at all 09:11 celeron55 appguru: you are now. let me know if you'd like to have helper as the main group which would change the color of your name 09:29 appguru celeron55: thanks, not necessary 09:29 appguru celeron55: thanks, not necessary 09:30 appguru my thunderbird seems to have broken, oof 09:35 appguru celeron55: BTW, now that we have ContentDB, I'd count CDB links as download links and not mandate a direct link to a .zip in the mod post. Thoughts? 09:37 appguru hmm, looks like I can't move stuff to "Old Games" 09:49 celeron55 eh i'll just make you a moderator instead 15:36 MTDiscord When trying to compile MT not with the system's luajit, but one I provide myself via LUA_INCLUDE_DIR and LUA_LIBRARY, cmake correctly says "Found LuaJIT: /home/myuser/temp/luajittest/LuaJIT/src/libluajit.a". But the very next line it claims "Using LuaJIT provided by system." How can I prove I am using the one "found" in the path I provided, not the one "provided by system"? 15:41 sfan5 locate the link.txt and check what it references 15:42 sfan5 I suspect the cmake message is just misleading 15:46 MTDiscord Ok, thanks. In that file there's /home/myuser/temp/luajittest/LuaJIT/src/libluajit.a so I should be good. Any way to prove it from ingame? 15:50 sfan5 no 15:52 MTDiscord Unfortunate, but so be it. Thanks! 15:56 MTDiscord Yes, the cmake message is half-wrong 16:00 fluxionary_ Bastrabun: if the version number of the system luajit is different than the one you're trying to link against, you should be able to check that 16:00 MTDiscord Unfortunately both say 2.1-beta3 16:10 fluxionary_ Bastrabun: try `ldd bin/minetest|grep luajit` 16:13 MTDiscord2 When I cmake this is what I get: You are using a relatively old version of LuaJIT. We recommend running a recent version (from git) as older ones are known not to build/work correctly in all cases. 16:16 MTDiscord2 When I then run the command suggested by flux it points to the system luajit 16:36 sfan5 it shouldn't 16:36 sfan5 but MT also can't link to both the system one and the libluajit.a you provided, so no idea 16:37 MTDiscord The order of those parameters is not important, right? 16:44 sfan5 which ones? 16:44 MTDiscord DLUA_LIBRARY, DLUA_INCLUDE_DIR and DENABLE_LUAJIT 16:45 sfan5 no 18:16 independent56 Hi. I have a docker image with Minetest running. It says it is listening on port 0.0.0.0:30000, but then my client times out when attempting to connect to 127.0.0.1:30000 despite te docker image being local. Even when i connect to what ip addr shows as a docker address, it times out 18:16 independent56 It says like "net 172.17.0.1/16" and then i connect to that and it times out 18:17 MTDiscord Did you map the port in your container to a port on the host machine? 18:17 independent56 ocker run --rm -p 30000:30000 -v /home/independent56/world:/world -v /home/independent56/logs:/log minetest 18:17 independent56 That's the command, and it does work 18:18 independent56 It runs the image perfectly but the port does not work 18:18 MTDiscord That -p might be assuming TCP 18:18 independent56 hmmm... EXPOSE 30000-31000/tcp is inside the dockerfile also 18:18 independent56 I'll go ask ChatGPT how to -p with UDP. 18:18 MTDiscord The expose doesn't actually DO anything. 18:19 independent56 So it's like a python docstring 18:20 independent56 It does DO things, but not anything practical on the computer level, just human level 18:20 independent56 And now it finally gives the error i CAN fix "name not registered" when i use -p 30000:30000/udp as per ChatGPT's advice 18:21 * independent56 experiences naeusa from lack of recent minetest experience 18:27 independent56 And now for a problem ChatGPT cannot solve. I am using CMD ["/opt/minetest/bin/minetestserver", "--world", "/world", "--port", "30000"] to run minetest. I am using -v ~/world:/world/ to create a bind mount to my home dir/world. However, i cannot see any data in ~/world. Also, the world is a new one whenever i start the docker container 18:28 independent56 Which means data is not being preserved 18:33 MTDiscord hmm, weird. I assume --world means world path and not world name ... I've generally only used world name and not path. 18:34 MTDiscord When you're using bind mounts with docker, keep in mind that user ID and permissions are preserved, so if you created the world dir as user 1000 but the process inside the container is running as e.g. 1001, then it probably won't actually have permission to write to the world. 18:34 independent56 i'll fix it 18:36 independent56 I ran chmod o+w world in my home directory. However, the docker still does not write to it 18:37 independent56 Minetest also says "ACTION[Main]: World at ['/world']" 18:37 independent56 And i'm using "--world", "/world" and the new flag -v /home/independent56/world:/world/ to publish a world at ~/world 19:02 kilbith user loss of interest: https://i.imgur.com/c6uPJk6.png 19:06 Desour weekend spike? 19:09 kilbith no, the download rate have collapsed (3x less) since the start of the feature freeze (coincidence?); I'm checking daily 19:11 Desour oh, I though the "(1 week)" means the time-span is 1 week. the feature freeze is already there for >1 week 19:12 rubenwardy It's the change in average score since a week ago 19:12 Desour it would be funny if mods get fewer downloads before a release because users fear that the new mod version will use new engine features and become incompatible 19:12 kilbith yes and I can see a durable loss of interest this month 19:13 rubenwardy no, total scroe 19:13 rubenwardy huh 19:13 rubenwardy thought it was average 19:13 rubenwardy delta(contentdb_score[7d]) 19:14 kilbith you better release something impressive quick 19:15 rubenwardy probably doesn't help that the android app is suspended 19:18 Desour rubenwardy: why does the avg package score have a sawtooth shape? are packages somehow building up score over a day? 19:19 rubenwardy yes, they get points from downloads and then lose 5% each day 19:19 rubenwardy see https://content.minetest.net/help/top_packages/ 19:20 Desour 👀 19:23 independent56 Fixed it! 19:23 independent56 Turns out worldname allows me to make a world called "world" 19:24 independent56 By changing the bind mount to go to container "/opt/minetest/worlds/world", i can now persist data 19:25 independent56 It's a hack, but if it works it works 19:25 kilbith I especially think that a lengthy feature freeze knocks out people's motivation to contribute 19:26 kilbith should be parallelized to another branch 19:27 kilbith you are killing the emulation 19:29 independent56 What feature freeze? 19:29 MTDiscord minetest 5.7.0 feature freeze 19:30 independent56 How did i not hear about this? Where can i learn more? 19:31 Desour independent56: i.e. here: https://dev.minetest.net/Git_Guidelines#Upstream_branch_rules 19:32 independent56 Sk nobody made an extra branch so nobody can contribute? 19:32 independent56 Shocking 19:33 Desour see also #13285 19:33 ShadowBot https://github.com/minetest/minetest/issues/13285 -- 🧊 Feature freeze time 19:34 Desour you can still contribute. the merge will just be delayed 19:34 kilbith master should never be freezed, that's the point 19:34 independent56 Oh so now people should wait until 5.8 for their features to be published? 19:37 cafblx I know some of you aren't on Discord, and I'm having issues with reading map blocks, so I'm cross-posting here: https://dpaste.com/8HWRDZ3NA 19:53 Desour cafblx: you can compare your code (and its output) with the engine. ppl won't be able to help you anyway if you don't show them your code 19:53 Desour besides that, note that node data is not compressed in your mapblock version 21:20 MinetestBot 02[git] 04sfan5 -> 03minetest/minetest: Update credits tab and mailmap 130a698d9 https://github.com/minetest/minetest/commit/0a698d92c3dc94057af9c4b83712593cf66c10eb (152023-03-29T21:19:11Z) 21:20 MinetestBot 02[git] 04sfan5 -> 03minetest/minetest: Bump protocol version 13bd88d29 https://github.com/minetest/minetest/commit/bd88d299b9159acbbf7d9d1df6d4f5487653a863 (152023-03-29T21:19:39Z) 22:12 MinetestBot 02[git] 04x2048 -> 03minetest/minetest: Return 'loops' occlusion culler under a setting (#13352) 139389895 https://github.com/minetest/minetest/commit/93898957b6c5862c6db0fa2555eaa5d3a0070e12 (152023-03-29T22:10:23Z) 23:46 fluxionary_ is there any reason why the lua api doc is .txt and not .md? i guess probably because that's where people and things expect it to be, and changing it would cause problems? 23:47 fluxionary_ would be nice if github would show the styled version