| Time |
Nick |
Message |
| 00:00 |
|
xerox123_ joined #minetest-hub |
| 00:02 |
|
nerzhul joined #minetest-hub |
| 00:05 |
nerzhul |
#8046 |
| 00:05 |
ShadowBot |
https://github.com/minetest/minetest/issues/8046 -- Fix player's attachments not updating visibility correctly by PiezoU005F |
| 00:25 |
|
ANAND joined #minetest-hub |
| 00:28 |
|
T4im joined #minetest-hub |
| 00:35 |
|
benrob0329 joined #minetest-hub |
| 01:13 |
|
sniper338 joined #minetest-hub |
| 02:22 |
|
Ruslan1 joined #minetest-hub |
| 03:17 |
|
MinetestSam joined #minetest-hub |
| 04:18 |
|
ANAND joined #minetest-hub |
| 04:42 |
|
ClobberXD joined #minetest-hub |
| 05:16 |
|
NathanS21 joined #minetest-hub |
| 07:00 |
|
ANAND joined #minetest-hub |
| 07:26 |
|
CWz joined #minetest-hub |
| 07:27 |
nerzhul |
#8058 |
| 07:27 |
ShadowBot |
nerzhul: Error: Delimiter not found in "HTTP Error 404: Not Found" |
| 09:00 |
|
jluc joined #minetest-hub |
| 09:08 |
|
jas_ joined #minetest-hub |
| 10:01 |
|
Krock joined #minetest-hub |
| 10:11 |
|
ANAND joined #minetest-hub |
| 11:30 |
|
_Xenon joined #minetest-hub |
| 11:46 |
|
Fixer joined #minetest-hub |
| 11:57 |
|
Krock joined #minetest-hub |
| 13:05 |
|
entuland joined #minetest-hub |
| 13:22 |
|
T4im joined #minetest-hub |
| 13:38 |
Krock |
weird flex but ok |
| 13:38 |
Krock |
"flex" is a very interesting CSS property. solves many of my tiling issues |
| 13:46 |
|
aerozoic joined #minetest-hub |
| 13:49 |
nerzhul |
Krock yeah i adopted it in many situations |
| 13:49 |
nerzhul |
but why here ? :p |
| 13:50 |
Krock |
had no better channels to go offtopic |
| 13:50 |
Krock |
thought to share it tho |
| 14:02 |
|
aerozoic_ joined #minetest-hub |
| 14:10 |
|
aerozoic__ joined #minetest-hub |
| 14:13 |
ANAND |
Lol |
| 14:37 |
|
kaeptmblaubaer joined #minetest-hub |
| 14:42 |
* rubenwardy |
really likes the idea of using Docker to build and provision MT server |
| 14:58 |
|
CWz joined #minetest-hub |
| 15:07 |
|
Gael-de-Sailly joined #minetest-hub |
| 15:38 |
|
cautiouspotato joined #minetest-hub |
| 16:25 |
|
cautiouspotato joined #minetest-hub |
| 16:40 |
celeron55 |
that's straight-up other worldly |
| 16:40 |
celeron55 |
i mean, the issue |
| 16:41 |
Krock |
people who read the logs will wonder about the gaps |
| 16:41 |
MinetestSam |
I just don't understand this line " you're a shame for all the humankind for being agains't blockchain" |
| 16:42 |
Krock |
that's double inverting |
| 16:46 |
BillyS |
Minetest authentication is complicated. |
| 16:47 |
Krock |
BillyS: in C++? oh yes it is |
| 16:47 |
BillyS |
I don't quite understand the salt business ... does the client generate the salt on the first authentication and then save it locally somehow for later authentications? |
| 16:48 |
BillyS |
I doubt anyone but the person who wrote this could answer that question. :P |
| 16:49 |
Krock |
if there's code it's possible to study it and answer that after a while. I thought the salt was based on the nickname |
| 16:49 |
Krock |
plus some magic value |
| 16:49 |
BillyS |
I've been studying this code for weeks |
| 16:50 |
BillyS |
I think the salt is based off the nick somehow, but I'm not quite sure how yet |
| 16:50 |
Krock |
BillyS: clientiface should contain most of the relevant code |
| 16:50 |
Krock |
mainly the init part |
| 16:50 |
BillyS |
Really? It seems that most of the authentication stuff is in client.cpp |
| 16:51 |
BillyS |
Oh, I guess there is some stuff about it there |
| 16:51 |
Krock |
nvm, you're right. clientiface is just a little bit of the auth stuff. I though there was more |
| 16:53 |
BillyS |
Hmm ... the docs say that it uses the SRP protocol as described here: https://tools.ietf.org/html/rfc2945 |
| 16:53 |
BillyS |
Yet the client seems to generate the salt, not the server |
| 16:56 |
rubenwardy |
loool |
| 16:58 |
|
Unarelith joined #minetest-hub |
| 17:06 |
BillyS |
Hump, tried the minetest wireshark plugin and it wouldn't even run |
| 17:07 |
BillyS |
Oh, I was using an older version. nvm |
| 17:10 |
|
MinetestSam joined #minetest-hub |
| 17:21 |
Fixer |
hi Krock |
| 17:21 |
Krock |
hi Fixer |
| 17:22 |
Fixer |
than* |
| 17:27 |
|
ANAND joined #minetest-hub |
| 17:52 |
|
scr267 joined #minetest-hub |
| 17:59 |
|
Ruslan1 joined #minetest-hub |
| 18:04 |
Fixer |
blockchain is dead, hype is dead, news at 11 |
| 18:04 |
Fixer |
so are shitcoins |
| 18:04 |
Fixer |
\o/ |
| 18:18 |
Fixer |
what the fuck https://github.com/minetest/minetest.github.io/issues/151 |
| 18:18 |
rubenwardy |
discussion in -dev |
| 19:03 |
Fixer |
does worldedit has undo? |
| 19:06 |
Krock |
https://github.com/Uberi/Minetest-WorldEdit/search?q=undo&unscoped_q=undo nope |
| 19:06 |
Krock |
the next thing you're going to say is: "shit" |
| 19:16 |
Fixer |
also, Back button is not reliable |
| 19:16 |
Fixer |
maybe I should update it again |
| 19:16 |
Fixer |
rubenwardy: what happens if I have two same modpacks, one from CDB and other from manual, will they conflict in some ways? |
| 19:17 |
Krock |
the mods conflict |
| 19:17 |
Krock |
not the modpacks |
| 19:20 |
Fixer |
http://doku.minetest.net/download I love the smell of necro in the morning |
| 19:20 |
Fixer |
why did not they DMCA god damned minetest.com? seriously :( |
| 19:21 |
Fixer |
kill it |
| 19:21 |
Fixer |
why it even exists? |
| 19:21 |
Fixer |
oss projects strive for hosting, and this outdated shit... from 2012 hosted... just for? wasting electricity? |
| 19:23 |
Krock |
Fixer: if you have a VPS for various programs it's not much of an issue to just run nginx forever |
| 19:25 |
celeron55 |
well i left it around in case it's useful; i think it had that subdomain as a backup since before it was replaced by the current site anyway |
| 19:25 |
celeron55 |
less work to not touch it |
| 19:26 |
Krock |
you can still set up redirects while keeping the files |
| 19:26 |
celeron55 |
that's work |
| 19:26 |
Krock |
so that people don't download these old binaries |
| 19:26 |
Krock |
especially now when it's the only "valid" download page on google |
| 19:27 |
celeron55 |
if it was a redirect, google would've removed it automatically along with the new one |
| 19:27 |
Krock |
hmm right |
| 19:27 |
celeron55 |
and probably will do that if i turn it into a redirect |
| 19:27 |
celeron55 |
i guess it could have a link to the current download page |
| 19:28 |
celeron55 |
so, maybe it's after all useful |
| 19:29 |
CWz |
who owns minetest.com |
| 19:29 |
celeron55 |
nobody knows |
| 19:29 |
Krock |
what if it redirects to minetest.net/ directly? |
| 19:29 |
Krock |
not necessary the downloads page |
| 19:29 |
entuland |
or redirecting it to a clone page at minetest.net/get-minetest or the alike may be an idea? |
| 19:30 |
Krock |
symlinks ftw |
| 19:51 |
|
Grandolf joined #minetest-hub |
| 19:52 |
Grandolf |
is there a bot i can use in irc for translating stuff automatically from spanish, french, german, and russian to english? |
| 19:53 |
Fixer |
yeah, redirect will be helpful |
| 19:57 |
Krock |
Grandolf: IhrFussel has such a feature |
| 19:57 |
Krock |
they use a web API afaik |
| 19:58 |
Grandolf |
yea but idk how any of that works |
| 19:58 |
Grandolf |
im a noob at fancy computer stuff |
| 19:58 |
Shara |
What Fussel has works in game, but it's not an IRC bot. |
| 19:59 |
Krock |
no, but the in-game mod translates all incoming messages |
| 19:59 |
Krock |
I think it would be rather simple to use the same API for an IRC bot |
| 19:59 |
Shara |
Looks like Grandolf wants somethign specifically for IRC though. |
| 20:00 |
Shara |
I think there was an IRC one already. I'm sure I saw it in use somewhere |
| 20:00 |
Shara |
lurk's bot maybe? |
| 20:00 |
Grandolf |
ill ask longer |
| 20:01 |
Shara |
He might have an idea. |
| 20:01 |
Shara |
Sorry I can't help more with it myself |
| 20:01 |
Grandolf |
its fine, ty |
| 20:23 |
nerzhul |
https://forum.minetest.net/viewtopic.php?f=18&t=21733 |
| 20:24 |
nerzhul |
!title |
| 20:24 |
MinetestBot |
nerzhul: Minetest 5.0.0 feature freeze ! - Minetest Forums |
| 20:24 |
Unarelith |
\o/ |
| 20:24 |
nerzhul |
can a moderator stick it ? |
| 20:24 |
nerzhul |
read, not just the title :p |
| 20:28 |
Unarelith |
I did, that came in early January for a release in early February, just as expected :D |
| 20:33 |
nerzhul |
later than i expected, but at least we are close now :) |
| 20:33 |
nerzhul |
translator will have some work to do |
| 20:47 |
|
IhrFussel joined #minetest-hub |
| 20:47 |
IhrFussel |
MTG has apple trees right? Or am I misremembering |
| 20:50 |
Shara |
Normal trees in MTG are apple trees |
| 20:50 |
Shara |
They just don't all grow apples |
| 20:51 |
IhrFussel |
I see so it's random |
| 20:52 |
IhrFussel |
And hi |
| 20:53 |
Shara |
Hi :) |
| 20:59 |
IhrFussel |
!seen shivajiva |
| 20:59 |
MinetestBot |
IhrFussel: shivajiva was last seen at 2019-01-01 11:56:02 UTC on #minetest-hub |
| 20:59 |
IhrFussel |
Weird...gone for 4 days ... isn't he usually here daily? |
| 21:05 |
Krock |
maybe in holidays? |
| 21:11 |
rdococ |
what's the best way to create a per-player inventory list with custom allow_put, allow_move etc.? |
| 21:12 |
rubenwardy |
5.0.0 and a player inventory |
| 21:12 |
rubenwardy |
or if in 0.4.x, a detached inventory |
| 21:13 |
rdococ |
I was planning to make this mod for 0.4.x but I'm probably going to switch to 5.0.0 |
| 21:18 |
rdococ |
is there a function to register a player inventory, or do I just create a new inventory list and use the "register_xxx_player_inventory_xxx" functions? |
| 21:19 |
rdococ |
s/second xxx/action |
| 21:22 |
rubenwardy |
the later |
| 21:22 |
rubenwardy |
registering a new player inventory would be silly |
| 21:26 |
rdococ |
ah |
| 21:57 |
rubenwardy |
sofar: you really should just delete VE's mods on minetest-mods https://forum.minetest.net/viewtopic.php?p=340064#p340064 |
| 21:57 |
rubenwardy |
you are breaking things because of this |
| 21:58 |
rubenwardy |
or at least use a bot to keep them in sink |
| 21:59 |
Shara |
A thousand times this ^ |
| 22:00 |
Krock |
I don't see how technic is related there but ok |
| 22:01 |
rubenwardy |
technic isn't maintained by vanessae iirc |
| 22:01 |
VanessaE |
rubenwardy: *I* deleted them. sofar disagreed. |
| 22:02 |
VanessaE |
and yeah, I don't touch technic, other than trivial PRs once in a while |
| 22:03 |
sofar |
well at this point I'll gladly do either (sync or delete) |
| 22:03 |
sofar |
so, what is preferred? |
| 22:03 |
sofar |
syncing isn't too difficult |
| 22:03 |
Krock |
convert to a mirror if possible |
| 22:04 |
VanessaE |
wtf, tl;dr |
| 22:04 |
Krock |
VanessaE: ditto |
| 22:04 |
rubenwardy |
can you push to archived repos? ie: repos with issues and PRs disabled |
| 22:05 |
sofar |
no idea |
| 22:05 |
sofar |
brb shower first, I spent 2 hours working outside, I smell |
| 22:05 |
rdococ |
okay now I'm doing something that causes minetest to quit without giving me an error |
| 22:10 |
rdococ |
amd I have no idea why |
| 22:11 |
entuland |
the thing you're doing is some mod you're coding, rdococ? |
| 22:12 |
rdococ |
yes |
| 22:13 |
IhrFussel |
WOW...I'm testing shivajiva's dev branch of sauth mod and max_lag is 0.15 with 8 players! |
| 22:13 |
entuland |
rdococ: the log shows nothing at all, it simply stops printing anything while the game exits? |
| 22:13 |
rubenwardy |
nice! |
| 22:14 |
IhrFussel |
The dev branch requires you to apply a .sql which moves the records into several tables like auth_abc auth_def etc |
| 22:14 |
rubenwardy |
IhrFussel: on production? |
| 22:14 |
rubenwardy |
any issues? |
| 22:14 |
IhrFussel |
None so far ... but I need to enable the profiler to see the real numbers |
| 22:15 |
IhrFussel |
Oh the dev sauth also implements a cache ... you can specify max amount of records to cache and which accounts to cache based on last login |
| 22:16 |
IhrFussel |
The options are |
| 22:16 |
IhrFussel |
sauth.cache_max = 500 -- maximum number of memory cached entries on startup |
| 22:16 |
IhrFussel |
sauth.cache_ttl = 86400 -- seconds deducted from last login |
| 22:17 |
rdococ |
nvm |
| 22:17 |
|
Fixer_ joined #minetest-hub |
| 22:17 |
entuland |
found the culprit, rdococ? |
| 22:18 |
rdococ |
was trying to do player:get_meta("field") instead of player:get_meta():get_int("field") |
| 22:18 |
entuland |
glad you found it |
| 22:19 |
* rubenwardy |
runs the migration script |
| 22:22 |
IhrFussel |
Wowowow... my server stays mostly at 0.15 sec max_lag now...before it was constantly 0.3-0.4 secs |
| 22:27 |
rubenwardy |
how long does it take max_lag to settle? |
| 22:27 |
rubenwardy |
CTF freezes for a few seconds at the beginning of a map, to place everything |
| 22:27 |
rubenwardy |
which throws it off a little |
| 22:28 |
IhrFussel |
IIRC it takes 5 minutes to be halfed from the highest max_lag IF the other steps were relatively non-delayed |
| 22:28 |
VanessaE |
Man, I was this -> <- close to telling that guy to piss-off. |
| 22:28 |
VanessaE |
https://forum.minetest.net/viewtopic.php?f=11&t=2041&p=340072#p340072 |
| 22:29 |
VanessaE |
blaming me because his google-fu sucks? |
| 22:30 |
VanessaE |
*growl* |
| 22:34 |
|
Unarelith joined #minetest-hub |
| 22:34 |
rubenwardy |
lool |
| 22:35 |
rubenwardy |
I've memorised mesecon's IDs on CDB because I use it for testing so much |
| 22:35 |
IhrFussel |
LOOK AT THIS BEAUTY |
| 22:35 |
IhrFussel |
- on_prejoinplayer[1] ............................... | 200 | 290 | 225 | 5.4 | 33.9 | 19.3 |
| 22:35 |
rubenwardy |
nice! |
| 22:35 |
Krock |
200us? |
| 22:35 |
rubenwardy |
ns |
| 22:36 |
rubenwardy |
iird |
| 22:36 |
rubenwardy |
wait no |
| 22:36 |
rubenwardy |
that can't be right |
| 22:36 |
Krock |
how? there's only get_time_us or so |
| 22:36 |
IhrFussel |
It's µs so 0.3 ms on average |
| 22:36 |
sofar |
the split to separate auth tables for abc/def sounds like a design problem, though |
| 22:36 |
sofar |
I really don't like that |
| 22:36 |
IhrFussel |
VS 150ms before |
| 22:36 |
Krock |
IhrFussel: did you update to 0.4.17.1? |
| 22:37 |
Krock |
changed the builtin code by any chance? |
| 22:37 |
IhrFussel |
I'm talking about sauth |
| 22:37 |
Krock |
and you were using it before as well? |
| 22:37 |
IhrFussel |
sauth lagged the server a lot |
| 22:37 |
IhrFussel |
Cause it only had 1 huge table |
| 22:37 |
Krock |
weird |
| 22:37 |
sofar |
I wonder what kind of queries sauth does in the first place that it takes so long |
| 22:37 |
IhrFussel |
And now the records are spread |
| 22:38 |
sofar |
there's always a cost |
| 22:39 |
IhrFussel |
Wait... those values might be incorrect cause I have cache enabled...it caches the last 1,000 accounts that last joined 2 days ago ... I have to disable it to test properly |
| 22:39 |
rubenwardy |
it was also because sauth didn't have an index on the usernames |
| 22:39 |
rubenwardy |
so it was doing a full table search each time |
| 22:40 |
sofar |
but that was fixed, right? |
| 22:40 |
rubenwardy |
in the dev version, which we're using right now |
| 22:40 |
rubenwardy |
the new version also has caching |
| 22:40 |
IhrFussel |
Well he fixed it now by having multiple tables |
| 22:41 |
sofar |
well that should go into master then, but multiple tables imho isn't a "fix" |
| 22:41 |
IhrFussel |
Here's the .sql https://github.com/shivajiva101/sauth/blob/dev/tools/schema_update.sql |
| 22:41 |
rubenwardy |
https://forum.minetest.net/viewtopic.php?t=20393 has some stats on the old db |
| 22:41 |
sofar |
I don't want the splitup, but I do want the index/cache |
| 22:41 |
Krock |
so how bad is the performance for the map then? there are surely more mapblock than auth records |
| 22:41 |
sofar |
those changes make actual sense |
| 22:41 |
rubenwardy |
it compares auth redux, sauth, builtin, ... |
| 22:42 |
sofar |
Krock: exactly |
| 22:42 |
IhrFussel |
I don't see the issue with having 11 tables |
| 22:42 |
rubenwardy |
it's hacky, but sqlite doesn't seem to have a way to partition by default anyway |
| 22:42 |
IhrFussel |
At least it's not one for each letter |
| 22:42 |
Krock |
rubenwardy: also stats for the old builtin method I believe (string concat) |
| 22:43 |
Krock |
but these are interesting stats - assuming the auth records were always the same |
| 22:44 |
rubenwardy |
sorcerykid does a lot of complex impressive stuff, but I'd wish she'd put her efforts elsewhere sometimes |
| 22:46 |
rubenwardy |
direct link: https://i.imgur.com/rnFkvmU.png |
| 22:46 |
rubenwardy |
to graph |
| 22:46 |
rubenwardy |
this is old sauth, mind |
| 22:49 |
IhrFussel |
shivajiva says he improved speed even without splitting -> 'I changed the table index to the name column and constrained it to ignore duplicates' |
| 22:49 |
IhrFussel |
But splitting improves it further |
| 22:50 |
rubenwardy |
that's what I meant by this <rubenwardy> it was also because sauth didn't have an index on the usernames |
| 22:50 |
rubenwardy |
<rubenwardy> so it was doing a full table search each time |
| 22:50 |
IhrFussel |
Yeah so now he fixed that and even WITHOUT caching I see avg µs of 250 |
| 22:50 |
IhrFussel |
Amazing |
| 22:53 |
IhrFussel |
sofar, multiple tables help definitely on servers with a huge amount of accounts (500,000+) but on smaller servers it's probably silly |
| 23:00 |
rubenwardy |
a disadvantage of the multiple table thing is you can't write simple SQL statements anymore |
| 23:01 |
rdococ |
Does swapping two item stacks count as two "move"s? |
| 23:01 |
rdococ |
(in the same item list ofc) |
| 23:01 |
rubenwardy |
one move, iirc |
| 23:01 |
rubenwardy |
I'm not sure |
| 23:07 |
|
T4im joined #minetest-hub |
| 23:08 |
IhrFussel |
rubenwardy, he wants to offer a simple .sql file which allows to at least purge old accounts based on time |
| 23:08 |
rubenwardy |
the update has given me a 83615x improvement |
| 23:08 |
IhrFussel |
So you don't have to go through each table manually |
| 23:08 |
rubenwardy |
bad statistics there though |
| 23:09 |
rubenwardy |
<rubenwardy> New profiler output: https://gist.github.com/rubenwardy/e478eb8f7af459be0cdddefa4ab11d04 |
| 23:09 |
rubenwardy |
<rubenwardy> This time sauth uses 0-1us, probably due to the caching or the server being quiet. This is 83615x faster! |
| 23:10 |
IhrFussel |
Try with caching false...it will almost be the same |
| 23:10 |
IhrFussel |
This is with false - on_prejoinplayer[1] ............................... | 2 | 4516 | 273 |
| 23:11 |
rubenwardy |
with true: - on_prejoinplayer[1] ............................... | 1 | 1 | 1 | 0.0 | 0.0 | 0.0 |
| 23:11 |
IhrFussel |
Well I'd say everything <1ms is...perfect |
| 23:12 |
rubenwardy |
the next sticking points for me are sban and chatplus |
| 23:15 |
IhrFussel |
Maybe he finds time to add indexing/splitting to sban too =O but we cannot expect that from him...he deserves a break |
| 23:17 |
* rubenwardy |
likes the idea of using docker to deploy Minetest servers |
| 23:18 |
IhrFussel |
Guess what? He just told me he's already onto sban! |
| 23:18 |
rubenwardy |
:O |
| 23:18 |
rubenwardy |
nice! |
| 23:18 |
rubenwardy |
<DELTA_FORCE> Why does everyone think I am a moderator there? |
| 23:18 |
rubenwardy |
<CTF_Bot> <paneg> cause you said you were one |
| 23:19 |
Krock |
I'm mod to |
| 23:19 |
Krock |
as in modder |
| 23:24 |
IhrFussel |
He says he would like his mods (sauth and sban) to stay one of the top choices for servers so he will continue to work on them when needed |
| 23:25 |
IhrFussel |
That is really nice considering I see a lot of modders just say 'i'm done with this mod, use it or don't...i don't care' |
| 23:31 |
rdococ |
uggg |
| 23:31 |
rdococ |
these calculations are really complicated |
| 23:31 |
rdococ |
I have several items, each of which has a "cost" - and you can equip those items to have their effects, but you can't go over a certain "cost" |
| 23:32 |
rdococ |
However, a "move" event could either swap the two stacks, or add to the "to" stack, and this depends on the type of the "to" stack and its count |
| 23:32 |
rdococ |
I'm considering just re-calculating the cost of the whole list at this point |
| 23:56 |
|
scr267 joined #minetest-hub |
