Time Nick Message 01:15 paramat this meta block update is a big issue, i approve of a fix even if it means dropping compatibility with older clients 01:15 paramat est31 thanks for merging waving stuff! 09:05 * T4im has 2 heisenbugs currently with minetest he can't report or reproduce well :| 09:06 T4im anyone else had that formspec flickering for example? 09:07 T4im seems to occur more likely when running minetest for longer periods of time with constant restarting, like you would do during testing/deving, (but attempting to reproduce it that way fails) 09:07 T4im restart solves it luckily, but still 09:08 rubenwardy video? 09:09 T4im I only get it, when I don't try to study/reproduce/observe it x) its not as much as a bad bug as one that drives one crazy, because its so evasive :D 09:12 rubenwardy How much is the flickering? 09:12 rubenwardy Do it's visibility completely change? 09:12 rubenwardy like, it flashes from 100% to 0% opacity 09:12 T4im becomes quite bad with time.. on-off-on-off-on-off 09:12 rubenwardy or does it move around 09:13 T4im it flickers between opened and closed 09:13 T4im or rather shown 09:13 rubenwardy when it's off, can you see the cross hair, etc 09:13 rubenwardy what mods are you using, which mod's formspec is ope? 09:13 T4im with formspecs like inventories and circsaw its also noticeable, that it happens to slow down the "listing" of inventory items... but only at initial loading of the formspec 09:14 T4im a consecutive one is better again 09:14 T4im mods are unimportant, its really "any" formspec then, server, locally, vanilla or not 09:14 rubenwardy so it occurs on vanilla? 09:15 T4im if it occurs, it occurs in the client completely.. even switching between multiplayer and singleplayer keeps that happening.. 09:16 T4im at first I thought it was caused by connecting/starting game sessions multiple times, either by a caching problem or by something not being freed.. but well.. heisenbug.. the moment you try to cause it, it doesn't happen anymore x) 09:16 T4im and starts again, when you already forgot about it again 09:21 T4im then again, wasn't there that serialization issue with the pascal strings? maybe something just writes randomly in the client around and causes my 2 heisenbugs as well 09:23 T4im the other heisenbug is worse btw: it causes nodes to be places into the wrong location.. like in pointed_thing.under instead of above.. it happens to all kinds of nodes, usually they just deny being placed, but with mods that use custom placement (homedecor expansions for example) but also fences for some weird reason, they eat "under" up 09:23 T4im that one was actually observed by other players as well though 09:24 T4im but is again, not really reproducable.. it happens regularly, but randomly 09:24 T4im I think more likely when teleporting or moving quickly into unloaded areas 09:24 T4im its really hard to reproduce in singleplayer.. only had it twice there 09:25 T4im sadly not yet with some useful debug statements :/ 09:25 T4im I don't believe it to be a mod bug though 09:27 rubenwardyy well, Formspec is Dead 09:27 T4im unless its also happening in builtin.. 09:27 T4im hmm haven't looked there yet 09:27 T4im yea.. the formspec thing is annoying, but solved by restart.. so not a biggy 09:27 rubenwardyy When did this start? 09:28 T4im noticed this a few weeks ago.. but I was inactive for half a year before.. 09:28 T4im so hard to say 09:37 T4im ok, the node eating issue seems to take contours... probably CONTENT_IGNORE, not properly tested in some places, mod issue to that degree.. the question remaining what content_ignore does in these places.. but now that I think of it, I've heared some players talk about how random content_ignore nodes where positioned in the world on occasion, and had that myself once too 09:37 T4im were* 09:38 T4im like.. you can actually move around them 09:39 T4im something worth to add to the smoketests :) 09:40 T4im also raising the question, why this happens to nodes without custom on_place 14:46 nrzkt celeron55: forum is out 14:48 rubenwardy did you see my comment here, hmmmm? #2610 14:48 ShadowBot https://github.com/minetest/minetest/issues/2610 -- Invalid block data in database: unsupported NameIdMapping version 14:53 rubenwardy does anyone have any insights as to why Minetest can only handle ~20 players at a time? 14:53 rubenwardy is it networking or environmental logic? 14:53 hmmmm rubenwardy, yes...? 14:54 hmmmm why does this apply to me in particular 14:54 celeron55 what the fuck happened to my server 14:56 rubenwardy oops, I misremembered, that's a different link 14:57 celeron55 ...what on earth could have taken it down like this 14:57 rubenwardy I thought that was a backtrace of https://github.com/minetest/minetest/blame/master/src/util/serialize.h#L265 14:57 rubenwardy but I got my issues mixed up 14:57 rubenwardy :S 14:58 rubenwardy What's up with it, celeron55? 15:03 celeron55 looks like the VPS got reset and openbsd is freaking out due to inconsistencies in the filesystem 8) 15:04 celeron55 and i can't run fsck because i can't type / because VNC screws up keyboard layouts 15:04 celeron55 well fuck 15:07 nrzkt oh you host it on OpenBSD, gg :D 15:07 nrzkt for preventing that i modified /etc/rc on sensible servers to add fsck -a -y :p 15:08 rubenwardyy do you mean "good good" 15:09 celeron55 now everything works again 15:10 celeron55 pro tip: tigervnc doesn't work and vinagre works if you aren't using an US keyboard locally 15:14 celeron55 nrzkt: i guess i'll do that if this reoccurs; this was the first time in the 105 days i've ran that server so i'm not too concerned about it right now 15:38 waressearcher2 celeron55: if someone wants to ask you a non-dev question and you are not in #minetest its impossible to ask you that question ? 15:41 celeron55 waressearcher2: uh, /msg exists 17:22 est31 paramat, protocol needs no incompatible change 17:22 est31 it needs a change yes, but no incompatible one 17:45 RealBadAngel est31, i will put there compability code, i do have almost evertything done, had not much time today to finish it 17:46 RealBadAngel but frankly i hate compability code at all 17:46 RealBadAngel thats sick imho 17:47 RealBadAngel not to mention that im fixing issue caused by compability code :P 17:47 RealBadAngel versions shall be strict and no gimmicks 19:08 est31 RealBadAngel, how is your pr plan 19:09 est31 do you want to make one big pr, with all the changes 19:09 est31 or do you want to separate them 19:09 est31 e.g. one for the sign drawtype change 19:09 est31 and one for the meta update per single node change 19:09 est31 its better to have it separate 19:13 est31 nrzkt, can you have a look at #3150 ? 19:13 ShadowBot https://github.com/minetest/minetest/issues/3150 -- Added more information about how get_node_light works. by RobertZenz 19:14 nrzkt est31: don't know how this lua call works but seems good 19:47 est31 pushing in 10 mins: https://github.com/est31/minetest/commit/183d0d5633b94bfa831528cb637fdb29342676d4 19:47 nrzkt thanks this will fix jenkins build 19:47 nrzkt because android build is broken without this patch, waiting is not needed. 19:48 est31 okay 19:48 est31 pushed 22:07 est Tesseract, it would be great if you could comment on #3159 22:07 ShadowBot https://github.com/minetest/minetest/issues/3159 -- Improve locale directory detection by est31 22:08 est it fixes my setup, where I have RUN_IN_PLACE=false, dont do make install, because I prefer package managers, and have worlds etc in ~/.minetest 22:30 red1 How do I reopen a pull request that was closed due to force push? 22:31 kahrl_ red1: I wasn't aware force push closes PRs... 22:31 kahrl_ issue #? 22:32 red1 game#667 22:32 ShadowBot https://github.com/minetest/minetest_game/issues/667 -- Add tnt API. by red-001 22:33 red1 The branch has new commits. 22:34 celeron55 someone from the core team has to reopen it 22:35 kahrl_ I've often force pushed my PRs and they were never closed by that 22:35 celeron55 it seems that that branch does not exist anymore 22:35 red1 It does https://github.com/red-001/minetest_game/tree/tnt-api 22:36 celeron55 oh 22:36 celeron55 so what happened is that you force-pushed master into it, and then pushed commits to it again? 22:36 red1 Yes 22:36 celeron55 then github saw that it had no actual commits and closed the PR 22:37 celeron55 i mean, in between 22:38 red1 Thanks 22:38 celeron55 i reopened it and github immediately refreshed your commits into it