Time |
Nick |
Message |
00:33 |
|
sagax joined #minetest |
00:39 |
|
fluxflux joined #minetest |
00:49 |
|
Hawk777 joined #minetest |
01:22 |
|
simon8162 joined #minetest |
01:25 |
|
xose[m] joined #minetest |
01:27 |
|
IronOxidizer[m] joined #minetest |
01:30 |
|
LPKI_Ramen joined #minetest |
01:30 |
|
tuxayo[m] joined #minetest |
01:33 |
|
silwol joined #minetest |
01:33 |
|
Kimapr[m] joined #minetest |
01:34 |
|
jfred[m]1 joined #minetest |
01:34 |
|
sturm joined #minetest |
01:36 |
|
rschulman joined #minetest |
01:37 |
|
FrostRanger[m] joined #minetest |
01:56 |
|
Pie-jacker875 joined #minetest |
02:03 |
|
pevernow joined #minetest |
02:05 |
pevernow |
I'm going to build a content DB mirror in China |
02:07 |
rubenwardy |
good luck |
02:07 |
rubenwardy |
easiest way would be to use nginx |
02:07 |
rubenwardy |
use a proxy and a proxy cache |
02:07 |
rubenwardy |
make the uploads have a super high timeout, and cache to disk |
02:08 |
rubenwardy |
you can even preload the cache using a script |
02:08 |
rubenwardy |
just make sure to proxy redirect rewrite to not bypass the cache |
02:10 |
pevernow |
That 's a good ideaBut does contentdb provide synchronization scheme with other contentdb? |
02:13 |
rubenwardy |
there's only one contentdb |
02:13 |
rubenwardy |
uploads are guaranteed to be uniquely named |
02:13 |
rubenwardy |
the API can be cached short-term |
02:13 |
rubenwardy |
ContentDB does not provide a facility to download everything |
02:14 |
rubenwardy |
you'll be ratelimited if you try to do that |
02:15 |
|
ShadowNinja joined #minetest |
02:15 |
|
hanetzer joined #minetest |
02:15 |
|
NathanS21 joined #minetest |
02:15 |
pevernow |
I think I'll have to try nginx proxy first. |
02:15 |
|
craigger joined #minetest |
02:15 |
|
greeter joined #minetest |
02:15 |
pevernow |
Thanks for your help |
02:15 |
|
Fusl joined #minetest |
02:16 |
|
Fulgen joined #minetest |
02:17 |
|
ircSparky joined #minetest |
02:18 |
|
Sires joined #minetest |
02:52 |
|
mdk joined #minetest |
03:11 |
|
FeXoR joined #minetest |
04:03 |
|
sivarajan joined #minetest |
05:09 |
|
sivarajan joined #minetest |
05:12 |
|
shangul joined #minetest |
06:10 |
|
MDude joined #minetest |
06:35 |
|
TomTom_ joined #minetest |
06:39 |
|
AndDT joined #minetest |
06:46 |
|
sagax joined #minetest |
07:06 |
|
Foz joined #minetest |
07:13 |
|
Flabb joined #minetest |
07:16 |
|
Sivarajan_R joined #minetest |
07:16 |
|
Out`Of`Control joined #minetest |
07:16 |
|
Menchers joined #minetest |
07:17 |
|
dabbill joined #minetest |
07:21 |
|
deltanedas joined #minetest |
07:33 |
|
deltanedas joined #minetest |
07:54 |
|
Foz joined #minetest |
07:55 |
|
Sivarajan_R joined #minetest |
08:00 |
|
ShadowNinja joined #minetest |
08:08 |
|
FeXoR joined #minetest |
08:19 |
|
Foz joined #minetest |
08:27 |
|
SwissalpS joined #minetest |
08:30 |
|
Sivarajan_R joined #minetest |
08:36 |
|
sivarajan joined #minetest |
08:47 |
|
deltanedas joined #minetest |
08:50 |
|
Sivarajan_R joined #minetest |
09:15 |
|
MTDiscord1 joined #minetest |
09:19 |
|
mizux joined #minetest |
09:28 |
|
calcul0n joined #minetest |
09:35 |
|
sivarajan joined #minetest |
09:38 |
|
Menchers joined #minetest |
10:03 |
|
Fixer joined #minetest |
10:22 |
|
Li0n joined #minetest |
10:23 |
|
jas_ joined #minetest |
10:28 |
|
deltanedas joined #minetest |
10:32 |
|
proller joined #minetest |
10:41 |
|
SX joined #minetest |
10:48 |
|
deltanedas joined #minetest |
11:16 |
|
AndDT joined #minetest |
11:27 |
|
deltanedas joined #minetest |
11:44 |
|
deltanedas joined #minetest |
12:15 |
|
bootlegbilly joined #minetest |
12:18 |
|
proller joined #minetest |
12:30 |
|
absurb joined #minetest |
12:41 |
|
sagax joined #minetest |
13:09 |
|
Jhalman joined #minetest |
13:13 |
|
deltanedas joined #minetest |
13:29 |
|
heavygale joined #minetest |
13:33 |
MTDiscord |
<Kajūto> Hey |
13:40 |
|
kamdard joined #minetest |
13:46 |
MTDiscord |
<HexBubble> Hey, Do you need anyhelp? |
13:49 |
|
deltanedas joined #minetest |
13:50 |
|
bootlegbilly joined #minetest |
14:03 |
|
Wuzzy joined #minetest |
14:07 |
|
dievri joined #minetest |
14:08 |
dievri |
Hello. Is settings from mod.conf already accessible after mod loaded, or I should manually load lite mod_settings = Settings("mod.conf") and after that just user mod_settings? |
14:08 |
rubenwardy |
yes, you'll need to use Settings() |
14:08 |
|
proller joined #minetest |
14:08 |
rubenwardy |
make sure to include get_modpath |
14:09 |
dievri |
rubenwardy, thanks! |
14:20 |
MTDiscord |
<oneplustwo> why is it you can float on air if you crouch on a block then break that block and continue holding shift? |
14:21 |
sfan5 |
you can't |
14:27 |
MTDiscord |
<oneplustwo> you can |
14:29 |
sfan5 |
I don't know about your settings or client source but I just opened a minetest game world and what your describe wasn't possible |
14:31 |
MTDiscord |
<oneplustwo> you literally hold shift, and break the block underneath you |
14:32 |
MTDiscord |
<Jonathon> what game? |
14:32 |
sfan5 |
I literally hold shift and break the node undearneath me, consequently I fall down |
14:33 |
|
bootlegbilly joined #minetest |
14:40 |
specing |
it's possible, but you have to be still partially standing on the neighboring block |
14:46 |
rubenwardy |
what Minetest version? This used to happen in 0.4 |
14:47 |
|
awell joined #minetest |
14:48 |
MTDiscord |
<oneplustwo> 5.3 |
15:06 |
|
Talkless joined #minetest |
15:19 |
jas_ |
specing: indeed you are correct |
15:22 |
specing |
Oh, I've always wondered... why the strange numbers for world limits? |
15:22 |
specing |
why 30912 or so, and not 2**15? |
15:25 |
MTDiscord |
<Warr1024> Probably safety margin |
15:25 |
MTDiscord |
<Warr1024> The closer you get to the actual limit, the more risk there is of being able to try to see into it, or having a really fast entity sort of "quantum tunnel" into that space, and it could cause weird bugs to shake out |
15:26 |
MTDiscord |
<Warr1024> The limit is 31000 in all directions, but I think the way it works a mapblock (or even mapchunk) needs to be within that limit to get generated. |
15:26 |
MTDiscord |
<Warr1024> Mapchunk alignment is ... weird. |
15:27 |
MTDiscord |
<Warr1024> I tried to make a mod to calculate what the ACTUAL map generation limits would be given the configuration, ran into all kinds of weird cases with very small limits, and ended up giving up and leaving it broken for now. |
15:34 |
specing |
so, why is the limit 31000 and not 32768 |
15:34 |
specing |
it's a large gap |
15:34 |
specing |
also, why is the limit 31000 and not 2**31? |
15:40 |
MTDiscord |
<Warr1024> large gap is because they wanted a large safety margin |
15:41 |
MTDiscord |
<Warr1024> 31k instead of 2G because that's the size of the integers they used for node coords |
15:41 |
MTDiscord |
<Warr1024> probably due to either irrlicht internal limitations or concerns about storage space |
15:41 |
MTDiscord |
<Warr1024> either way, there has been a lot of talk about raising the limits but nobody is actually doing the work necessary |
15:42 |
MTDiscord |
<Warr1024> which, btw, is theorized to be quite a lot. |
15:43 |
MinetestBot |
[git] Wuzzy2 -> minetest/minetest_game: Add icons for invisible nodes (#2718) 6605724 https://git.io/JUDQN (2020-09-30T15:41:04Z) |
15:45 |
specing |
About storage: how much of the storage is consumed by chunks that were explored once and then noone returned to them? |
15:45 |
specing |
could probably save a lot of chunks that were unvisited for months were just pruned away |
15:45 |
specing |
could probably save a lot of space by removing* chunks that were unvisited for months were just pruned away |
15:45 |
specing |
ugh that sentence |
15:45 |
specing |
let's try again |
15:46 |
specing |
could probably save a lot of space by removing chunks that were unvisited for months |
15:47 |
MTDiscord |
<Warr1024> I actually have a mod that collects per-mapblock statistics, including stuff like how many times players have dug/placed nodes in them. The original idea was to be able to prune such mapblocks. I just never got around to doing the work on the pruning part. |
15:58 |
|
deltanedas joined #minetest |
16:05 |
specing |
Warr1024 can you share any interesting insights? |
16:06 |
MTDiscord |
<Warr1024> The main insight I've gained from it is the fact that I apparently often end up gathering a lot of data that I never make the time to analyze. |
16:13 |
specing |
:( |
16:24 |
MTDiscord |
<Warr1024> https://gitlab.com/sztest/szutilpack/-/tree/master/szutil_usagesurvey if you're interested. It's also on CDB as part of szutilpack. |
16:56 |
|
Jhalman joined #minetest |
17:02 |
|
deltanedas joined #minetest |
17:30 |
|
fruitsnack joined #minetest |
17:41 |
|
FeXoR joined #minetest |
17:51 |
|
homthack joined #minetest |
18:06 |
|
calcul0n_ joined #minetest |
18:28 |
|
dievri joined #minetest |
18:29 |
dievri |
Hello. in mod.conf I have very long string in format key=value,value,value,value,value,value ..... an so on. How can I spread this on a multiple lines? |
18:30 |
dievri |
btw, there is also a Discord channel of minetest? |
18:33 |
MTDiscord |
<Jordach> https://discord.gg/6W84ytH |
18:35 |
MTDiscord |
<dievri> thanks! |
18:46 |
|
swift110 joined #minetest |
18:47 |
|
aldum joined #minetest |
18:49 |
Krock |
dievri: setting = """ |
18:49 |
Krock |
multi |
18:49 |
Krock |
line setting |
18:49 |
Krock |
""" |
18:49 |
Krock |
and that's documented in lua_api.txt |
18:51 |
Krock |
though this will add newlines at the end of each line, hence ensure that it works properly in your case |
18:51 |
rubenwardy |
depends etc ignores whitespace, and just looks for commas |
18:52 |
rubenwardy |
I think that's the only comma-seperated field? |
18:52 |
|
fluxflux joined #minetest |
18:52 |
Krock |
mapgen flags,"initial_give_items": "Am I a joke to you?" |
18:53 |
Krock |
also mod security (overall and HTTP-API) |
18:53 |
|
Fixer_ joined #minetest |
18:54 |
rubenwardy |
ahh, I was thinking of mod.conf |
18:54 |
rubenwardy |
well, they did say mod.conf |
18:57 |
Krock |
oh right |
18:58 |
Krock |
that part was already off-screen |
18:59 |
|
Fixer joined #minetest |
19:00 |
|
dievri joined #minetest |
19:12 |
|
Zughy joined #minetest |
19:36 |
|
_Zaizen_ joined #minetest |
20:02 |
|
proller joined #minetest |
20:03 |
|
erlehmann joined #minetest |
21:21 |
clavi |
should I change "max_packets_per_iteration" on a server, or is 1024 fine? |
21:25 |
sfan5 |
how many players? |
21:26 |
sfan5 |
did you get relevant warning(s) in your log? |
21:26 |
|
fruitsnack joined #minetest |
21:27 |
clavi |
5 players at most, but perhaps on crappy connections |
21:27 |
clavi |
had a pathological case rn with an american player on bad wifi (server is in germany) |
21:27 |
clavi |
they had "WARNING[ConnectionSend]: con(16/4) Packet quota used up for peer_id=1, was 1024 pkts" in their logs |
21:28 |
sfan5 |
you don't need to increase it then |
21:29 |
sfan5 |
especially not if that message pops up in the *clients* log |
21:30 |
clavi |
okay |
21:30 |
clavi |
the server itself has a fast connection, but I can't do anything about players far away on not-so-great internet |
21:31 |
clavi |
could I increase it for a LAN-only server though? and to what value? |
21:31 |
clavi |
ethernet lan that is, with at least 1 gbps |
21:34 |
sfan5 |
you can theoretically increase it to a very large value in a LAN but I'm pretty sure part of the reason for the limit is also to not overwhelm clients |
21:35 |
sfan5 |
(very large = 10000 or 65335 if you want, doubt it makes a difference at that point) |
22:12 |
|
proller joined #minetest |
22:20 |
|
deltanedas joined #minetest |
22:26 |
est31 |
emphasis on LAN here, wifi is different |
22:26 |
|
Menchers joined #minetest |
22:26 |
est31 |
it'd just try to retransmit the same huge packet over and over |
22:27 |
est31 |
wait it's numbers of packets |
22:27 |
est31 |
not packet sizes |
23:27 |
|
fluxflux joined #minetest |
23:42 |
|
homthack joined #minetest |