| Time |
Nick |
Message |
| 00:00 |
|
ssieb joined #minetest-hub |
| 00:07 |
|
paramat joined #minetest-hub |
| 00:09 |
|
red-001 joined #minetest-hub |
| 00:15 |
|
Dargod joined #minetest-hub |
| 03:42 |
|
Mr_Pardison joined #minetest-hub |
| 03:42 |
Mr_Pardison |
greetings! |
| 04:17 |
|
Corvus_ joined #minetest-hub |
| 04:22 |
|
BakerPrime_ joined #minetest-hub |
| 04:23 |
|
BakerPrime joined #minetest-hub |
| 05:24 |
|
behalebabo joined #minetest-hub |
| 06:39 |
|
paramat joined #minetest-hub |
| 06:42 |
paramat |
!tell IhrFussel do you know very roughly the MT version when your 'randomly stopping vehicles' issue started? did it happen in 0.4.16 stable? |
| 06:42 |
MinetestBot |
paramat: I'll pass that on when IhrFussel is around |
| 07:19 |
|
Corvus__ joined #minetest-hub |
| 07:30 |
|
twoelk joined #minetest-hub |
| 07:42 |
|
CWz joined #minetest-hub |
| 08:22 |
|
DI3HARD139 joined #minetest-hub |
| 09:11 |
|
ManElevation joined #minetest-hub |
| 09:14 |
|
jluc joined #minetest-hub |
| 09:32 |
|
IhrFussel joined #minetest-hub |
| 09:32 |
MinetestBot |
IhrFussel: Jul-20 06:42 UTC <paramat> do you know very roughly the MT version when your 'randomly stopping vehicles' issue started? did it happen in 0.4.16 stable? |
| 09:34 |
IhrFussel |
!tell paramat 99% sure it did not happen in 0.4.16 but in backport 0.4.17 from Dec ... and it didn't happen in 0.4.17.1 still so I think it is fixed there...did you test 0.4.17.1? |
| 09:34 |
MinetestBot |
IhrFussel: yeah, sure, whatever |
| 10:18 |
IhrFussel |
The stalling issue makes my server partly unplayable... I advise every server owners with an active playerbase to check the chat log and search for things like "laaaag" or "map not loading" or similar |
| 10:19 |
|
tenplus1 joined #minetest-hub |
| 10:20 |
tenplus1 |
hi folks :) |
| 10:21 |
IhrFussel |
The stalling does not ONLY affect mapblock sending |
| 10:22 |
tenplus1 |
also affects inventory movement (from yesterday) |
| 10:22 |
tenplus1 |
hi fussel |
| 10:22 |
IhrFussel |
Or well...it does technically I guess...mesecons won't work when it happens |
| 10:25 |
IhrFussel |
Ihave 10 videos for proof now that clearly show that only certain app users with random names cause this...the map loads again as soon as the problem user logs off |
| 10:26 |
tenplus1 |
thing is... is it the area they are in causing issue, sauth or some player data being loaded/saved ?? |
| 10:26 |
IhrFussel |
There is nothing busy server side |
| 10:27 |
IhrFussel |
Everything else works, the clients somehow BLOCK map sending/updating |
| 10:27 |
tenplus1 |
so something about that particular client is stalling mapload ? |
| 10:27 |
IhrFussel |
One client is somehow able to ruin the game for everyone else |
| 10:29 |
tenplus1 |
you could always set: strict_protocol_version_checking = true |
| 10:29 |
tenplus1 |
and force players to use 0.4.17 and above |
| 10:31 |
rubenwardy |
FFS, why are longs 64 bits in the kernel |
| 10:31 |
rubenwardy |
This is silly |
| 10:31 |
tenplus1 |
hi ruben, also dunno what that is |
| 10:31 |
tenplus1 |
*means |
| 10:31 |
IhrFussel |
First I need to make sure that it affects all MT servers and not just mine (which is very unlikely since the hardware of the server is pretty modern) |
| 10:32 |
rubenwardy |
Longs are usually 32bit, and is that size when in other code in Android |
| 10:33 |
rubenwardy |
This is why u64 is a good idea when the size matters |
| 10:33 |
tenplus1 |
how does it affect things with longs 64bit |
| 10:34 |
rubenwardy |
It's used as flags to control a function. A bit like CSM flavour limits, instead of using tons of booleans you use a number and set individual bits |
| 10:34 |
tenplus1 |
ohhh |
| 10:34 |
tenplus1 |
gotcha |
| 10:35 |
IhrFussel |
tenplus1, do you save the chat log of users? In case you do could you please look for certain messages that could indicate that map loading issue? |
| 10:36 |
tenplus1 |
no-one has ever complained about map issues on Xanadu, my inbox only contains books with coords for banned/limited players protected areas |
| 10:36 |
tenplus1 |
but then again, we're still on 0.4.16 |
| 10:39 |
IhrFussel |
It happened before 0.4.17.1 already ... but I meant the chat log of people chatting publicly |
| 10:39 |
tenplus1 |
it's monitored, yes... but like I said no-one has ever mentioned issues with the map pausing/loading/glitching |
| 10:43 |
IhrFussel |
It is definitely *not* all app users... not all random names cause this, but quite a few |
| 10:43 |
tenplus1 |
coudl the particular player that causes the issue be using csm mods to stress information between server and client ? |
| 10:45 |
IhrFussel |
Too many different people for that |
| 10:45 |
IhrFussel |
I'm pretty sure it is a broken app that handles mapblocks wrong |
| 10:46 |
|
lumberJ joined #minetest-hub |
| 10:46 |
tenplus1 |
so many bogus clients that could do many weird things.. we need a way to stop them connecting |
| 10:46 |
tenplus1 |
hi lumber |
| 10:46 |
IhrFussel |
But ONE client should never be able to stall the ENTIRE map |
| 10:46 |
tenplus1 |
yeah, it's annoying |
| 10:47 |
IhrFussel |
So maybe the server waits for each client to acknowledge something and never gets a reply or something |
| 10:47 |
tenplus1 |
true, we need a dev to ask for that one |
| 10:48 |
IhrFussel |
Or some kind of endless loop between server <-> client |
| 10:48 |
IhrFussel |
I could reduce block sends per client to see if that fixes/improves it |
| 10:49 |
tenplus1 |
worth a try |
| 10:49 |
IhrFussel |
It was 20, after restart 10 |
| 10:50 |
IhrFussel |
What's the value on xanadu? |
| 10:50 |
tenplus1 |
checking |
| 10:51 |
tenplus1 |
max_simultaneous_block_sends_per_client = 3 |
| 10:51 |
tenplus1 |
max_simultaneous_block_sends_server_total = 40 |
| 10:53 |
IhrFussel |
I think the latter setting is ignored, not used anymore by the engine |
| 10:54 |
tenplus1 |
:P |
| 10:54 |
IhrFussel |
Only 3? Map loads quick with that value? |
| 10:54 |
tenplus1 |
surprisingly yes |
| 10:54 |
tenplus1 |
I can use fast and run any direction and map loads quick enough to handle it |
| 10:54 |
IhrFussel |
I think setting it higher means less network packages |
| 10:55 |
tenplus1 |
considering server is in US and I'm in UK |
| 10:58 |
IcyDiamond |
i wonder if its possible to make an npc "build" a schematic |
| 10:58 |
IcyDiamond |
aka place nodes in sequence |
| 10:58 |
IcyDiamond |
instead of just spawning the schematic |
| 10:58 |
tenplus1 |
you could have a schematic saved as a table, loop through table from bottom to top and use npc/mob pathfinding to move to position and place node and keep going |
| 10:59 |
IcyDiamond |
i guess |
| 10:59 |
IcyDiamond |
i might try at some point |
| 10:59 |
tenplus1 |
also hi :))) hi raven |
| 11:01 |
IcyDiamond |
lol hi |
| 11:01 |
IcyDiamond |
:P |
| 11:02 |
Raven262 |
Hi tenplus1 |
| 11:03 |
tenplus1 |
o/ |
| 11:05 |
IhrFussel |
First symptom of the map stalling is map loading very slow suddenly |
| 11:05 |
tenplus1 |
and it only happens when a specific player signs on ??? |
| 11:06 |
IcyDiamond |
rubenwardy: https://i.lunasqu.ee/Screenshot_18-07-20-14:05:33.png is it just me or is the api doc unreadable now? |
| 11:06 |
IhrFussel |
And works again when that specific player leaves yep |
| 11:06 |
tenplus1 |
and youv'e asked that user what client they use ??? |
| 11:06 |
IhrFussel |
Well the map doesn't stall right away...it can take a moment |
| 11:07 |
IhrFussel |
I can try but most app users don't understand English |
| 11:08 |
rubenwardy |
IcyDiamond: any errors in the console? |
| 11:08 |
|
Fixer joined #minetest-hub |
| 11:08 |
tenplus1 |
o/ fixer |
| 11:08 |
rubenwardy |
What os/gtk theme? |
| 11:09 |
Fixer |
hi |
| 11:09 |
IcyDiamond |
rubenwardy: the links on the sidebar all 404 as well |
| 11:09 |
rubenwardy |
Must have broke some links then |
| 11:09 |
rubenwardy |
What's the URL? |
| 11:09 |
IcyDiamond |
https://rubenwardy.com/minetest_modding_book/lua_api.html |
| 11:14 |
IhrFussel |
tenplus1, any reason why minetest.item_eat(5) won't work in the beverage mod? I use stamina as hunger mod |
| 11:14 |
IhrFussel |
It just won't fill the bar when I drink something |
| 11:15 |
rubenwardy |
Code? |
| 11:16 |
IcyDiamond |
tenplus1: is pathfinding in the engine or your mobs mod? |
| 11:16 |
|
ManElevation joined #minetest-hub |
| 11:16 |
IhrFussel |
https://github.com/minetest-mods/beverage/blob/master/init.lua#L136 |
| 11:16 |
IhrFussel |
In on_use() |
| 11:17 |
IhrFussel |
Wait... |
| 11:18 |
IhrFussel |
Why is it a node and not a craft item? |
| 11:20 |
tenplus1 |
fussel, if you use minetest.item_eat(value) it shoudl work fine |
| 11:20 |
IhrFussel |
Didn't know you can (ab)use nodes as tools for drinking |
| 11:20 |
|
Krock joined #minetest-hub |
| 11:20 |
Krock |
hi tenplus1 |
| 11:21 |
tenplus1 |
IcyDiamond: minetest.find_path exists but in the actual mob api we have pathfinding as well to move mobs |
| 11:22 |
Krock |
sup? |
| 11:22 |
IcyDiamond |
ok, i think i'll use your mobs mod |
| 11:22 |
tenplus1 |
fussel: I use nodes for wine mod: https://notabug.org/TenPlus1/wine/src/master/init.lua#L44 and wine is eaten/drank ok with stamina mod |
| 11:22 |
IhrFussel |
That is so weird, I hear the drinking sound (defined before minetest.item_eat()) but no stamina change |
| 11:23 |
Krock |
hi IhrFussel, IcyDiamond :) |
| 11:23 |
IhrFussel |
Hey Krock |
| 11:24 |
IcyDiamond |
hey Krock |
| 11:24 |
|
lumberJ joined #minetest-hub |
| 11:24 |
tenplus1 |
icy, setting self.path to hold minetest.find_path coords starts them off, also make sure pathfinding = 1, is set in mob def |
| 11:25 |
tenplus1 |
hi Krock |
| 11:25 |
Krock |
heh tenplus1. 5 minute delay this time :D |
| 11:25 |
Krock |
you must be totally busy |
| 11:26 |
tenplus1 |
hehe sorry dude, was looking through chat.. paused scroll so I coudl catch up |
| 11:27 |
tenplus1 |
o.O ouch, the code is kinda wonky in beverage mod |
| 11:27 |
tenplus1 |
all spaced wrong in places |
| 11:29 |
IhrFussel |
It's a rather old mod but I need some drinks :P |
| 11:29 |
tenplus1 |
heh |
| 11:29 |
tenplus1 |
farming redo has some added, but yeah... need to add more drinks... how about wine mod ? :PPPPP |
| 11:30 |
IhrFussel |
my players want tea and other healthy stuff =P |
| 11:30 |
tenplus1 |
nice |
| 11:30 |
IhrFussel |
I also have the sandwiches mod by Annalysa and that one works |
| 11:31 |
IhrFussel |
And the sandwiches also just use on_use = minetest.item_eat(5), |
| 11:32 |
tenplus1 |
are you using thirty mod ? |
| 11:32 |
IhrFussel |
Nope |
| 11:33 |
tenplus1 |
in init.lua, line 160, have it say: return minetest.item_eat(5) and see if that helps |
| 11:35 |
tenplus1 |
same with line 201 |
| 11:36 |
tenplus1 |
return minetest.item_eat(5) |
| 11:37 |
twoelk |
wasn't there some problem with certain mobile clients granting a too far interact reach? maybe that triggers too much map load |
| 11:38 |
IcyDiamond |
rubenwardy: so.. what's up with the docs? |
| 11:38 |
IcyDiamond |
error on my side? |
| 11:41 |
rubenwardy |
It's using the static CSS of my main site, as the page is missing a relative URL root |
| 11:41 |
|
aerozoic joined #minetest-hub |
| 11:41 |
tenplus1 |
hi aerozoic |
| 11:41 |
aerozoic |
wazup tenplus1! |
| 11:42 |
tenplus1 |
tinkering... always tinkering :PP you ? |
| 11:43 |
rubenwardy |
IcyDiamond: https://github.com/rubenwardy/minetest_modding_book/commit/b4057fda023d2c2189ef5333caa9c98aacba37a3 |
| 11:44 |
rubenwardy |
IcyDiamond: seen the new chapters? |
| 11:47 |
IcyDiamond |
2018-07-20 14:46:45: ERROR[Main]: .../evert/minetest/bin/../mods/entitybuildqueue/example.lua:4: calling 'place_schematic_on_vmanip' on bad self (userdata expected, got table) |
| 11:47 |
IcyDiamond |
i wish it told me which arg |
| 11:47 |
rubenwardy |
You're probably giving it data rather than vm |
| 11:48 |
IcyDiamond |
VoxelManip() returns data and not a vm? |
| 11:48 |
rubenwardy |
Returns a vm |
| 11:48 |
IcyDiamond |
then whats the issue? |
| 11:48 |
rubenwardy |
You then get data with :get_data() |
| 11:48 |
rubenwardy |
Right |
| 11:48 |
rubenwardy |
Idk then |
| 11:48 |
rubenwardy |
Code? ;) |
| 11:49 |
IcyDiamond |
VoxelManip() returns nil |
| 11:50 |
IcyDiamond |
but the docs state that the position args are optional? |
| 11:50 |
IcyDiamond |
why does it return nil then |
| 11:50 |
tenplus1 |
you could have npc change self.state to something unrecognised by api "building" for instance so the npc does nothing, then use do_custom to face direction and place node with punch animation and keep doing that until schem file built... then finally change self.state to "stand" |
| 11:51 |
IcyDiamond |
it returns nil even with arguments |
| 11:51 |
IcyDiamond |
im doing something wrong again |
| 11:51 |
IcyDiamond |
lmao |
| 11:52 |
rubenwardy |
When are you calling it? |
| 11:52 |
IcyDiamond |
that matters? |
| 11:52 |
IcyDiamond |
can't call it at load time? |
| 11:52 |
IcyDiamond |
so i guess i'll have to put it in an after()? |
| 11:52 |
IcyDiamond |
¯\_(ツ)_/¯ |
| 11:53 |
IcyDiamond |
nope still nil |
| 11:53 |
IcyDiamond |
well im confused |
| 11:55 |
IcyDiamond |
can i get an example of place_schematic_on_vmanip please |
| 12:09 |
|
lumberJ joined #minetest-hub |
| 12:33 |
IcyDiamond |
i dont understand why VoxelManip returns nil |
| 12:34 |
tenplus1 |
make sure any coords are rounded , doesnt seem to work with things like 32.5 etc |
| 12:34 |
IcyDiamond |
https://i.lunasqu.ee/Screenshot_18-07-20-15:34:16.png |
| 12:34 |
tenplus1 |
ahh |
| 12:36 |
IcyDiamond |
it doesnt say anywhere when or where it should be called |
| 12:38 |
tenplus1 |
https://dev.minetest.net/minetest.place_schematic_on_vmanip |
| 12:38 |
IcyDiamond |
do i even need vm to get data out of a schematic? |
| 12:38 |
tenplus1 |
yeh |
| 12:38 |
tenplus1 |
check example link |
| 12:38 |
IcyDiamond |
I saw that already |
| 12:39 |
Krock |
how about write_to_map? |
| 12:39 |
IcyDiamond |
? |
| 12:40 |
IcyDiamond |
i need the data from a schematic, not place it in the world |
| 12:44 |
Krock |
local placed = minetest.place_schematic_on_vmanip(...) |
| 12:44 |
Krock |
if that returns false, then you must specify the positions |
| 12:44 |
IcyDiamond |
minetest.get_voxel_manip() returns nil |
| 12:45 |
Krock |
err what |
| 12:45 |
IcyDiamond |
i have the same question |
| 12:45 |
Krock |
https://github.com/minetest/minetest/blob/master/src/script/lua_api/l_env.cpp#L963-L976 there's no way it can return nil |
| 12:46 |
IcyDiamond |
then what the f is happening here |
| 12:47 |
Krock |
here it works just fine https://github.com/SmallJoker/peoplecantlua/blob/master/peoplecantblur/init.lua#L127 |
| 12:48 |
IcyDiamond |
i am super confused |
| 12:50 |
xerox123_ |
Hello all o/ |
| 12:50 |
tenplus1 |
hi xerox |
| 12:50 |
rubenwardy |
IcyDiamond: at load time, there is no world |
| 12:50 |
rubenwardy |
The environment is loaded after mods load |
| 12:51 |
rubenwardy |
As you can't know all node definitions until loading is finished |
| 12:51 |
tenplus1 |
use a minetest.after function to wait 3 seconds then place schem file |
| 12:51 |
Krock |
lol. you're calling this on load? |
| 12:51 |
rubenwardy |
Ah |
| 12:52 |
rubenwardy |
It'll be setting a nil table |
| 12:52 |
rubenwardy |
Nil meta tabld |
| 12:52 |
rubenwardy |
So it doesn't return nil, but instead an empty table |
| 12:52 |
IcyDiamond |
hmm what |
| 12:52 |
IcyDiamond |
ok i must be going crazy |
| 12:52 |
IcyDiamond |
it didnt work before with minetest.after either, now it does |
| 12:52 |
IcyDiamond |
well then |
| 12:52 |
rubenwardy |
L |
| 12:52 |
rubenwardy |
Ol |
| 12:53 |
IcyDiamond |
i have the confusions |
| 12:56 |
IcyDiamond |
static int ScriptApiBase::luaPanic(lua_State*): A fatal error occured: LUA PANIC: unprotected error in call to Lua API (not enough memory) |
| 12:56 |
IcyDiamond |
haha lol |
| 12:57 |
rubenwardy |
Oh dear |
| 12:57 |
rubenwardy |
That's... Not good |
| 12:57 |
IcyDiamond |
i tried to dump the data |
| 12:57 |
IcyDiamond |
that didn't go well |
| 12:57 |
IcyDiamond |
XD |
| 12:57 |
rubenwardy |
LVM exposes lots of weirdness like this when done wrong |
| 12:57 |
rubenwardy |
Oh wait lol |
| 12:57 |
rubenwardy |
Is that a puc version of an oom? |
| 12:58 |
IcyDiamond |
2018-07-20 15:56:11: ERROR[Main]: ServerError: AsyncErr: environment_Step: OOM error from mod 'entitybuildqueue' in callback environment_Step(): not enough memory |
| 12:58 |
IcyDiamond |
XD |
| 12:59 |
IcyDiamond |
how tf am i supposed to know how big of a vm i have to load |
| 12:59 |
IcyDiamond |
or create |
| 12:59 |
IcyDiamond |
place_schematic_on_vmanip requires me to specify positions |
| 13:00 |
tenplus1 |
its one command i've never used |
| 13:00 |
tenplus1 |
prefer normal schematic placement... |
| 13:00 |
tenplus1 |
if anything I'd only use vmanip to load the area in question then place normal schem |
| 13:01 |
IcyDiamond |
its when you want to do the crazy shiz you run into problems |
| 13:01 |
IcyDiamond |
XD |
| 13:01 |
rubenwardy |
Store the schem size in a meta file |
| 13:01 |
IcyDiamond |
dammit |
| 13:01 |
rubenwardy |
Also, if you can work out how to emerge a schematic in an lvm with correct lighting |
| 13:01 |
rubenwardy |
Let me know |
| 13:01 |
rubenwardy |
Lol |
| 13:01 |
IcyDiamond |
lmao |
| 13:02 |
rubenwardy |
Or, alternatively find out how to read schem size from the MTS file |
| 13:02 |
IcyDiamond |
too much work |
| 13:02 |
IcyDiamond |
lol |
| 13:02 |
rubenwardy |
Lol |
| 13:03 |
Krock |
[LOL INTENSIFIES] |
| 13:04 |
tenplus1 |
::P |
| 13:04 |
IcyDiamond |
Motivation rate drops to 0% |
| 13:04 |
Krock |
Suicide rate rises to 100% |
| 13:04 |
IcyDiamond |
I don't feel like killing myself over a game |
| 13:04 |
IcyDiamond |
¯\_(ツ)_/¯ |
| 13:08 |
tenplus1 |
is this just to place a schematic, or the start of the npc building thing |
| 13:08 |
rubenwardy |
CTF has separate confs for this |
| 13:09 |
rubenwardy |
ItB reimplements schematics using LVMs, and supports meta in the conf |
| 13:19 |
IhrFussel |
Can a core dev confirm that max block sends total is an obsolete setting? |
| 13:29 |
sfan5 |
"obsolete"? |
| 13:29 |
tenplus1 |
deprecated |
| 13:33 |
sfan5 |
if it was obsolete it would not exist anymore |
| 13:33 |
tenplus1 |
lol |
| 13:33 |
sfan5 |
which is the case in 5.0, but not in 0.4.17 |
| 13:34 |
Fixer |
https://twitter.com/without_pity/status/1019611100323680258 |
| 13:35 |
tenplus1 |
lol |
| 13:38 |
IhrFussel |
sfan5, could the removal of that setting cause such map stalling issues? |
| 13:39 |
sfan5 |
you're not running 5.0 are you? |
| 13:42 |
IhrFussel |
No, but I removed the block sends total from my .conf because I was told that it#s not used anymore |
| 13:43 |
sfan5 |
you should add that setting back |
| 13:44 |
IcyDiamond |
!title |
| 13:44 |
MinetestBot |
IcyDiamond: Without Pity op Twitter: "Its 33°C in Finland, inside the article circle.… " |
| 13:45 |
IcyDiamond |
lol same here and i live in estonia, which is just south of finland |
| 14:04 |
Krock |
IhrFussel: for reference, it's still in 0.4.x https://github.com/minetest/minetest/blob/stable-0.4/minetest.conf.example#L908 |
| 14:04 |
Krock |
it's removed in 5.0-dev, which was already mentioned earlier |
| 14:07 |
IhrFussel |
Yes but if it's missing it sets the default (40) no? So it shouldn't be related to the map sending issues |
| 14:08 |
Krock |
40 is quite low for popular servers, but it does indeed take the default value if the setting is not specified |
| 14:09 |
tenplus1 |
xanadu is set to 40 |
| 14:10 |
Krock |
also the settings do not change in-game (exception is probably "/set"), so they surely can't be an issue here |
| 14:15 |
IhrFussel |
And no dev got an idea how a "bad client" could "block" map from sending to other clients? Is there a security hole? Does the client request too many blocks at once? |
| 14:19 |
IhrFussel |
It looks like the more "bad clients" play the sooner it stalls ... if only 1 is connected with such a client it still loads but very slow |
| 14:20 |
Krock |
to find out more, you'd have to add more stats to the minetest binary -> custom build |
| 14:21 |
Krock |
or analyse the network traffic (but the packets won't be easily readable) |
| 14:21 |
|
jluc joined #minetest-hub |
| 14:24 |
rubenwardy |
I had a similar issue |
| 14:24 |
rubenwardy |
Although I didn't notice anything about clients |
| 14:25 |
rubenwardy |
As CTF is perpetually full |
| 14:25 |
rubenwardy |
I did a wireshark |
| 14:25 |
rubenwardy |
Didn't really gain much though |
| 14:33 |
|
ManElevation joined #minetest-hub |
| 15:26 |
|
orbitaldecay joined #minetest-hub |
| 15:27 |
orbitaldecay |
Greetings! |
| 15:29 |
Krock |
hi orbitaldecay |
| 15:41 |
|
AndroBuilder joined #minetest-hub |
| 15:42 |
|
aerozoic joined #minetest-hub |
| 15:51 |
|
ManElevation_ joined #minetest-hub |
| 16:11 |
|
entuland joined #minetest-hub |
| 16:31 |
Krock |
rubenwardy: "Moreover, admin shall not merge PRs in mods until a reasonable time has elapsed, and author has been given a reasonable time to review or object to a PR." ~ https://minetest-mods.github.io/README.html |
| 16:31 |
Krock |
so what's a reasonable time? two weeks? |
| 16:31 |
Krock |
a month? or multiple months? |
| 16:31 |
rubenwardy |
I don't know |
| 16:31 |
rubenwardy |
sofar? |
| 16:32 |
tenplus1 |
200 minutes |
| 16:32 |
rubenwardy |
I've been naughty and just merged PRs before |
| 16:32 |
Krock |
lol tenplus1. That's clearly too short |
| 16:32 |
tenplus1 |
oops, 220 minutes then |
| 16:32 |
Krock |
rubenwardy: me too on bugfix PRs which have a good quality |
| 16:32 |
tenplus1 |
falling.lua will need to be looked at soon enough and fixed |
| 16:32 |
rubenwardy |
I added screenshot.png to mesecons like this |
| 16:33 |
Krock |
very naughty indeed. lol |
| 16:58 |
|
benrob0329 joined #minetest-hub |
| 16:58 |
tenplus1 |
o/ benrob |
| 16:59 |
benrob0329 |
o/ tenplus1 |
| 17:05 |
|
garywhite joined #minetest-hub |
| 17:05 |
tenplus1 |
hi gary |
| 17:06 |
garywhite |
hi ten |
| 17:07 |
tenplus1 |
you got any servers running benrob ? |
| 17:09 |
benrob0329 |
No, I've thought about it a few times though |
| 17:09 |
tenplus1 |
heh :) |
| 17:09 |
tenplus1 |
am looking for some testers for a new falling.lua replacement |
| 17:12 |
tenplus1 |
any takers ? |
| 17:14 |
|
rdococ joined #minetest-hub |
| 17:16 |
tenplus1 |
wb rdococ |
| 17:17 |
Krock |
ok, let's give it a try. < tenplus1 |
| 17:17 |
tenplus1 |
https://pastebin.com/XBpYT62E <-- make an empty mod with only "default" in depends.txt .. it'll override original entity for falling nodes with new functions |
| 17:17 |
|
jluc joined #minetest-hub |
| 17:18 |
Krock |
uhm.. "local builtin_shared = ..." I'm sceptical |
| 17:19 |
|
Gael-de-Sailly joined #minetest-hub |
| 17:19 |
tenplus1 |
heh, ignore that line |
| 17:19 |
tenplus1 |
hi Gael-de-Sailly |
| 17:19 |
Krock |
"if below_node.name == "ignore" then" in combination with "get_node" might also delete them half-way falling |
| 17:22 |
tenplus1 |
that does same as original falling.lua, just removes entities from unloaded areas |
| 17:22 |
Krock |
ehh |
| 17:22 |
tenplus1 |
yeh, that was added |
| 17:22 |
tenplus1 |
if falling node drops into unloaded map area it's removed |
| 17:23 |
tenplus1 |
was meant for edges of map |
| 17:23 |
tenplus1 |
dropping sand off the side kinda thing :D |
| 17:24 |
Krock |
that could be solved properly using get_node_or_nil |
| 17:24 |
Krock |
when it returns a valid "ignore" node then it's the world bottom |
| 17:24 |
Krock |
or world edge |
| 17:25 |
tenplus1 |
the main difference with this is that falling nodes no longer replace smaller nodeboxes like before... that was a serious glitch... it checks first to make sure it's buildable_to or attached_node and can drop as item isntead of setting as node |
| 17:30 |
sofar |
rubenwardy: Krock: 30 days is the cutoff, but I often merge things 10-14 days old for repos that I know people are using and there's no sense in delaying patches getting merged |
| 17:31 |
tenplus1 |
o/ sofar |
| 17:32 |
sofar |
heya |
| 17:32 |
Krock |
I see, thanks sofar |
| 17:33 |
sofar |
one of the reasons I don't like to define numbers for words like "reasonable" is that it'll make people just complain if you merge something 1 day too fast |
| 17:38 |
Krock |
on the other hand it's not sure whether weeks or months are maent |
| 17:38 |
Krock |
*meant |
| 17:39 |
sofar |
ah |
| 17:41 |
Krock |
tenplus1: node_fall_hurt doesn't seem to work (not defined in minetest.conf) |
| 17:41 |
tenplus1 |
if you drop one of the falling nodes registered at top above you it should land and cause you damage |
| 17:41 |
twoelk |
reasonable time to respond may be longer during holiday seasons :-D |
| 17:42 |
Krock |
and same to the builtin falling node, six falling nodes suddenly become 4 by digging the buttom one |
| 17:42 |
tenplus1 |
oh, that's due to the small timer I added to simulate lag |
| 17:42 |
tenplus1 |
set delay to 0 |
| 17:42 |
tenplus1 |
line 34 |
| 17:44 |
Krock |
oh I see. there's no falling_node_damage group |
| 17:44 |
tenplus1 |
there is, yes... the very top function just adds it to sands and tnt :) |
| 17:45 |
* twoelk |
is off to the weekend |
| 17:45 |
Krock |
..and desert sand, which I use right now to test |
| 17:45 |
tenplus1 |
but once the changes are complete I could make a pull and add the new group to .conf and api txt |
| 17:45 |
|
twoelk left #minetest-hub |
| 17:45 |
* tenplus1 |
dropped sand on his head and died... only had 1 heart :D |
| 17:48 |
|
Gael-de-Sailly joined #minetest-hub |
| 17:48 |
Krock |
dropping it on my head. nothing happens |
| 17:49 |
tenplus1 |
you using 0.5 ? |
| 17:49 |
tenplus1 |
or 0.4.17 ? |
| 17:49 |
Krock |
no, 5.0-dev :P |
| 17:49 |
tenplus1 |
ahh, I've not even tested for that yet |
| 17:49 |
tenplus1 |
try dropping sand on mobs, works ok... new player model might be iffy |
| 17:50 |
Krock |
yes, you'd have to check 1m further down |
| 17:50 |
tenplus1 |
tried the get_node_or_nil instead of ignore and got a ton of "2018-07-20 18:50:14: WARNING[Main]: Map::removeNodeMetadata(): Block not found" errors |
| 17:51 |
Krock |
is that the client or server? |
| 17:51 |
tenplus1 |
client |
| 17:51 |
Krock |
aha |
| 17:51 |
tenplus1 |
? |
| 17:51 |
Krock |
PR coming soon (TM) |
| 17:52 |
tenplus1 |
local below_node = core.get_node_or_nil(below_pos) ; if not below_node then remove and return end |
| 17:52 |
tenplus1 |
will go back to normal get_node and checking for "ignore" for now |
| 17:53 |
Krock |
not remove. return. |
| 17:54 |
Krock |
so that the entity can be saved into the mapblock |
| 17:54 |
IhrFussel |
tenplus1, stamina.eat() doesn't even get called with the drinks |
| 17:55 |
IhrFussel |
I added log messages to that function but it doesn't log anything |
| 17:56 |
IhrFussel |
If I return minetest.item_eat() the itemstack will not be returned right? |
| 17:56 |
tenplus1 |
check this fussel, will make it work: https://pastebin.com/vUdig3Cn |
| 17:58 |
tenplus1 |
Krock: but if sand dropped off edge of world wont that just have it constantly there and not be removed ? |
| 17:58 |
Krock |
no, it'll be loaded again |
| 17:58 |
tenplus1 |
trying |
| 17:58 |
Krock |
and the on_step code sets the acceleration again |
| 17:59 |
tenplus1 |
nope, once placed off the side of the world it stays there never to be removed... would rather it be deleted |
| 18:14 |
IcyDiamond |
tenplus1: do mobs have a unique id? |
| 18:14 |
IcyDiamond |
that i could access from self |
| 18:14 |
tenplus1 |
only traders |
| 18:14 |
tenplus1 |
but no, never needed a unique id for them before |
| 18:15 |
|
longerstaff13 joined #minetest-hub |
| 18:15 |
|
longerstaff13 joined #minetest-hub |
| 18:15 |
IcyDiamond |
ok |
| 18:16 |
IcyDiamond |
oh wait, self can have custom vars |
| 18:16 |
tenplus1 |
yup, you can easily add your own |
| 18:16 |
IcyDiamond |
ok this makes it easier |
| 18:16 |
tenplus1 |
e.g. self.id = (math.random(1, 1000) * math.random(1, 10000)) .. self.name .. (math.random(1, 1000) ^ 2) |
| 18:17 |
tenplus1 |
this is what traders use |
| 18:17 |
IcyDiamond |
yeah im going to use it for build queues too |
| 18:18 |
tenplus1 |
remember that all self.* values are stored inside mob entity, so if it's in the middle of doing something (like building) it'll remember when it spawns in again |
| 18:19 |
IcyDiamond |
def |
| 18:19 |
IcyDiamond |
? |
| 18:19 |
tenplus1 |
huh ? |
| 18:19 |
IcyDiamond |
mob entity as any instance of it? |
| 18:19 |
IcyDiamond |
or what |
| 18:20 |
tenplus1 |
yup, you can spawn 20 npc's and assign 1 to build or do something with do_custom and that one npc will remember and continue |
| 18:20 |
IcyDiamond |
ok |
| 18:22 |
tenplus1 |
it does reset self.state to "stand" when they re-appear, but if you have self.path ready or self.build or whaverer you can check against it and set to "build" maybe to handle that |
| 18:22 |
Fixer |
CHAPA |
| 18:56 |
|
Gael-de-Sailly joined #minetest-hub |
| 18:59 |
IcyDiamond |
tenplus1: how do i get the position of a mob |
| 19:00 |
tenplus1 |
local pos = self.object:get_pos() |
| 19:00 |
IcyDiamond |
ok ty |
| 19:02 |
IcyDiamond |
tenplus1: so, can i make a mob walk to a specific position somehow too? |
| 19:02 |
tenplus1 |
if mob has self.pathfind = 1 in it's definition you can set self.path = minetest.find_path() to guide it somewhere and it'll follow |
| 19:03 |
IcyDiamond |
oh okay, cool |
| 19:03 |
tenplus1 |
sand monsters and npc have pathfind enabled by default :) |
| 19:07 |
Krock |
tenplus1: still trying to reproduce the client metadata warning |
| 19:09 |
tenplus1 |
Krock: change line:158 to read: local below_node = core.get_node_or_nil(below_pos) |
| 19:10 |
tenplus1 |
and line:158 to read: if not below_node then |
| 19:10 |
tenplus1 |
and drop a sand block off the edge of the world... that produces the error |
| 19:11 |
rubenwardy |
Yessss |
| 19:11 |
tenplus1 |
on 0.4.17-1 at least :) dunno about your 0.5 |
| 19:11 |
rubenwardy |
Soteria caught an apple cheatern |
| 19:12 |
Krock |
if not below_node then ... I assume "return end"? |
| 19:14 |
tenplus1 |
does sand drop ok and land/set ok ? andif you drop on things like grass/flowers it replaces them ok ? and buildable_to items like torches drop ok ? |
| 19:16 |
|
Gael-de-Sailly joined #minetest-hub |
| 19:17 |
IcyDiamond |
2018-07-20 22:16:21: ERROR[Main]: ServerError: AsyncErr: ServerThread::run Lua: Runtime error from mod 'npcbuildqueue' in callback luaentity_run_simple_callback(): Mod security: Blocked attempted read from house.meta |
| 19:17 |
IcyDiamond |
aw you cant read a file thats in the mod files? |
| 19:17 |
Krock |
tenplus1: dropped a node at -30910 to the bottom side of the world. no warning |
| 19:18 |
tenplus1 |
weird, I get pages worth of errors when I do it on 0.4.17-1 |
| 19:18 |
Krock |
IcyDiamond: it's because Minetest does not know that it's your mod |
| 19:18 |
Krock |
it would work on startup |
| 19:18 |
IcyDiamond |
dammit |
| 19:19 |
rubenwardy |
How does CTF do this? |
| 19:19 |
IcyDiamond |
so much for having a metadata file for a schematic... |
| 19:19 |
rubenwardy |
Is Settings exempt? |
| 19:19 |
Krock |
tenplus1: strange. could you please send me the exact file that you used to cause it? |
| 19:19 |
IcyDiamond |
now im completely screwed |
| 19:19 |
rubenwardy |
IcyDiamond: it works with Setting |
| 19:19 |
IcyDiamond |
it is a setting.. |
| 19:19 |
rubenwardy |
The lua class for reading .conf files |
| 19:19 |
IcyDiamond |
2018-07-20 22:17:56: ERROR[Main]: [C]: in function 'Settings' |
| 19:19 |
IcyDiamond |
does it require the extension? |
| 19:20 |
Krock |
probably |
| 19:20 |
IcyDiamond |
nope, still errors |
| 19:20 |
tenplus1 |
Krock: this is the code that causes the error: https://pastebin.com/U7w61T2m |
| 19:21 |
Krock |
and you're simply dropping sand towawrds the world edge? |
| 19:21 |
rubenwardy |
IcyDiamond: look at CTF in ctf_maps/schem_map |
| 19:21 |
IcyDiamond |
oh wait nvm |
| 19:22 |
rubenwardy |
It reads conf during run time |
| 19:22 |
IcyDiamond |
my mistake |
| 19:22 |
IcyDiamond |
i forgot to prepend the mod path |
| 19:22 |
rubenwardy |
It also has dynamic map reloading |
| 19:22 |
tenplus1 |
yeah, built a little tower at the edge and drop sand off the edge... ton of errors |
| 19:22 |
rubenwardy |
So I can add maps without restarting |
| 19:23 |
Krock |
it replaces buildable_to nodes nicely, that's all I found out |
| 19:24 |
Krock |
no single warning/error |
| 19:24 |
tenplus1 |
and if it drops on a thin nodebox the falling block will drop as an item isntead |
| 19:24 |
tenplus1 |
that was the glitch that players would use to remove teleport pads/protectors etc |
| 19:28 |
Krock |
2018-07-20 21:28:21: WARNING[Server]: Undeclared global variable "np" accessed at /data/Minetest/run/bin/../mods/tenplus1_falling/init.lua:235 |
| 19:28 |
Krock |
that's all I get |
| 19:28 |
|
jluc joined #minetest-hub |
| 19:28 |
tenplus1 |
checking |
| 19:29 |
tenplus1 |
oops, np shoudl be npos for sound... ehehe,, oops |
| 19:34 |
tenplus1 |
could you try placing a very thin node like a carpet (something not replacable) and place tnt far above it, light and see if ti disappears ok |
| 19:35 |
IcyDiamond |
hm why does my find_path return nil |
| 19:35 |
tenplus1 |
that means it cannot find path |
| 19:35 |
IcyDiamond |
its literally a straight line |
| 19:35 |
IcyDiamond |
minetest.find_path(pos, pos2, 16, 3, 3, "A*") |
| 19:36 |
tenplus1 |
use "Dijkstra" instead f A* |
| 19:36 |
IcyDiamond |
tried it, still nil |
| 19:36 |
tenplus1 |
so weird |
| 19:36 |
IcyDiamond |
indeed |
| 19:37 |
tenplus1 |
and pos1 and pos2 are ok ? |
| 19:37 |
IcyDiamond |
yes |
| 19:41 |
Sokomine |
the pathfinder from burli is quite nice and gives good results |
| 19:41 |
tenplus1 |
you got the link for that Sokomine |
| 19:42 |
tenplus1 |
keep forgetting to check it out :D |
| 19:42 |
Sokomine |
https://forum.minetest.net/viewtopic.php?f=9&t=17406 i use it (at least one version) in my mob_world_interaction mod |
| 19:43 |
Sokomine |
probably not so intresting for your mobs, ten. they can't spend much time on calculating a good path and probably shouldn't learn how to open doors :-) |
| 19:43 |
tenplus1 |
if mob pathfinding is set to 2 then they tend to break down walls and doors anyhoo :D |
| 19:45 |
Sokomine |
buuh! no breaking down of walls |
| 19:45 |
Sokomine |
rather let them build their own houses |
| 19:45 |
tenplus1 |
would be nice tow atch them build |
| 19:58 |
tenplus1 |
hrm, they dont seem to go through the doors when using pathfinder |
| 19:59 |
IcyDiamond |
tenplus1: what exatly is "per tick" |
| 20:00 |
tenplus1 |
per tick, every 0.01 second |
| 20:00 |
tenplus1 |
so 100 ticks per second |
| 20:00 |
IcyDiamond |
hm ok |
| 20:00 |
tenplus1 |
if you do something with do_custom it runs per tick, if you want it to run every 1 second you need to add a timer to handle that |
| 20:01 |
tenplus1 |
check out rat.lua in mobs animal for example |
| 20:01 |
Sokomine |
tenplus1: no, the pathfinder can just be modified so that it can find paths through doors. my mob_world_interaction mod is intended to help humanoid mobs who want to open doors and gates |
| 20:02 |
tenplus1 |
ahh ok |
| 20:03 |
tenplus1 |
cause mod kinda gave up with pathfinder, but redo api has it climbing on the roof to fall into a hole to get player |
| 20:07 |
IcyDiamond |
tenplus1: nothing is happening |
| 20:07 |
IcyDiamond |
or its taking a while cuz dtime is always like 0.03 |
| 20:09 |
tenplus1 |
show me code icy |
| 20:09 |
IcyDiamond |
oh wait |
| 20:09 |
IcyDiamond |
nvm |
| 20:09 |
tenplus1 |
:P |
| 20:10 |
Krock |
sofar: do you remember whether this code worked reliable to test? https://github.com/minetest/minetest/issues/5246#issue-208040774 |
| 20:22 |
Krock |
wrote a mod to update the node each second with a new rotation (set_node) and yet no warnings |
| 20:26 |
IcyDiamond |
can i make entity look at a position |
| 20:26 |
IcyDiamond |
or rotate towards |
| 20:27 |
Krock |
yes. |
| 20:28 |
Krock |
local dir = vecor.direction(pos1, pos2) entity.object:set_yaw(math.atan2(-dir.x, dir.z)) |
| 20:29 |
Krock |
if it's the wrong rotation, add another 90° or change the atan2 function accordingly |
| 20:29 |
tenplus1 |
if you mean a mob then use mobs:yaw(self, yaw0 |
| 20:30 |
tenplus1 |
if you mean a mob then use mobs:yaw(self, yaw( |
| 20:30 |
tenplus1 |
then it'll smooth rotate to look that direction |
| 20:31 |
tenplus1 |
nite folks :) |
| 20:31 |
|
tenplus1 left #minetest-hub |
| 20:32 |
|
paramat joined #minetest-hub |
| 20:32 |
MinetestBot |
paramat: Jul-20 09:34 UTC <IhrFussel> 99% sure it did not happen in 0.4.16 but in backport 0.4.17 from Dec ... and it didn't happen in 0.4.17.1 still so I think it is fixed there...did you test 0.4.17.1? |
| 20:46 |
|
ManElevation joined #minetest-hub |
| 20:52 |
|
benrob0329 joined #minetest-hub |
| 21:04 |
sofar |
Krock: never found a reliable reproducer, so no |
| 21:05 |
Krock |
too bad. :/ |
| 21:29 |
Calinou |
https://www.xkcd.com/583/ |
| 21:44 |
Krock |
tHEy'rE dOInG a bAtTLe r0ya1E iN MiN3tEsT oH gAwD https://forum.minetest.net/viewtopic.php?f=10&t=20528& |
| 21:45 |
Krock |
tHE meMeS arRiVEd |
| 21:45 |
* Sokomine |
takes a bucket of water and pours it over krock (obviously a case of too much heat) |
| 21:46 |
Krock |
tHaNK YoU bUt ThAT dIdn'T s0LV3 tHE iS5U3 |
| 21:49 |
* Sokomine |
wonders weather a bucket of lava might help :-) |
| 21:49 |
* Krock |
RuNs aWAy qUiCklY |
| 21:50 |
Fixer |
Krock: I SEE A PROBLEMS WITH YOUR FONTS< IT IS A BUG IN GTK RENDERING FROM TWOTHOUSANDS>>> |
| 21:53 |
IcyDiamond |
https://www.youtube.com/watch?v=14YCZ0aA68I |
| 21:53 |
IcyDiamond |
first vid on my channel lol |
| 21:53 |
|
Fixer_ joined #minetest-hub |
| 21:54 |
Krock |
congRaTuLaTIonS, ICydiAMoNd |
| 21:54 |
IcyDiamond |
XD |
| 21:54 |
Krock |
you seem to be an arch user btw |
| 21:55 |
IcyDiamond |
you dont say |
| 21:56 |
Krock |
i toTaLlY DOn'T SAY anYThInG |
| 21:57 |
IcyDiamond |
XD |
| 22:04 |
Krock |
IcyDiamond: could it be that your Minetest binary needs updating? I see those lighting issues on the ground |
| 22:05 |
IcyDiamond |
Probably |
| 22:06 |
IcyDiamond |
I updated today though |
| 22:08 |
Krock |
oh, that's recent enough |
| 22:09 |
Krock |
whereas the npc movement is nice, they seem to place nodes outside the usual range |
| 22:15 |
paramat |
i guess they're simply pathfinding towards the currently placed node |
| 22:16 |
IcyDiamond |
The movement is quite crap, I'll have to experiment with it further |
| 22:26 |
paramat |
it's not so bad :) |
| 22:51 |
|
Corvus_ joined #minetest-hub |
| 23:51 |
|
Tmanyo joined #minetest-hub |
