Time Nick Message 01:40 sofar # /usr/local/bin/minetestserver --version 01:40 sofar /usr/local/bin/minetestserver: /lib64/libstdc++.so.6: version `CXXABI_1.3.8' not found (required by /usr/local/bin/minetestserver) 01:40 sofar what the ... 01:42 * sofar ldconfigs 01:42 sofar # /usr/local/bin/minetestserver --version 01:42 sofar Minetest 5.0.0 (Linux) 01:42 sofar finally, got a mt-5.0.0 on centos7 build 01:42 sofar (never again) 01:44 Emerald2 Now you can nuke centos from orbit? :p 01:47 sofar if you uninstall minetest-server on centos, it removes the critically needed file in /etc/ 01:47 sofar this is why we can't have nice things 01:47 sofar (if only I could install clearlinux on this host, but I can't) 01:48 GreenDimond Ill stick with my Ubuntu 01:48 sofar not for a server 01:48 GreenDimond Why not 01:50 sofar haha 01:50 sofar you're funny 01:51 GreenDimond I use Ubuntu for my server 01:51 GreenDimond fight me 01:51 sofar no need 01:51 GreenDimond whats wrong with using ubuntu for a server 01:53 sofar well for one, you're leaving an enormous amount of performance potential on the table 02:17 rubenwardy I doubt it 03:06 paramat Emerald2 "more info about how the noise params in mapgen effect the generated terrain?" there are short descriptions in conf.example https://github.com/minetest/minetest/blob/0b492f82f77eb121d0c9af66ddf5e86f3ebbc876/minetest.conf.example#L2086 . also, for mgv7, somewhere in https://forum.minetest.net/viewtopic.php?f=47&t=15272 03:09 Emerald2 Thanks paramat. I'll have a look at those. 03:11 paramat ideally i *should* write detailed docs for all mapgens ... 03:17 Emerald2 It's pretty complex. Like, if I change this number to be bigger or smaller, what happens? 03:18 Emerald2 Like your 'to do' list isn't already a mile long. 03:59 paramat yes, it's important to understand how the noise params affect the behaviour of the resulting noise value, i've covered that in lua_api.txt. the 2nd step is to understand how that noise value affects an aspect of mapgen 04:00 paramat it's very difficult to make the first step easy for someone but i've tried 12:29 jas_ looooooooool 12:29 jas_ you should help him 12:29 jas_ no? 12:30 IhrFussel I already tried to tell him that his mods are most likely broken 12:30 IhrFussel And then he started with this whole 'giving up on Mt soon cause nobody cares about his server/malicious clients' blah blah 12:32 IhrFussel Now in his most recent post he says 'devs could find a way to block ad-infested/modified clients but they don't want to cause they probably like it the way it is' 12:33 IhrFussel And that 'open source' is no excuse ... by definition open source means anyone can see exactly how the code is written and that makes it automatically possible to fake everything the server ever asks from the client 12:33 IhrFussel But he doesn't understand that 12:35 IhrFussel Yes the devs could add a 'user agent' like thing into the client...but as soon as the mobile app devs would find out that their missing agent is rejected by servers they would simply add it to their version 12:35 IhrFussel It's useless to even try 12:47 BuckarooBanzai hi all 12:58 Krock hi BuckarooBanzai 12:59 BuckarooBanzai Testing pandorabox on v5.0, looks good so far :D 12:59 Krock o/ 13:03 rubenwardy BuckarooBanzai: yay 13:09 IhrFussel hi Krock 13:09 Krock hi IhrFussel 13:11 IhrFussel Since restart (after applying my solution) the highest time was this: CRAFT TIMER: Steuber336 moved 5 default:steel_ingot to craft grid (82573 µs) 13:12 IhrFussel At least far away from 200+ms 13:16 Krock and in plain u_i it was up to a second? 13:22 Fixer CHAPPAAAA 13:22 IhrFussel Without the inv:is_empty("craft") check it went up to 500ms...sometimes even 800 13:23 IhrFussel One sec, making a small list to compare before/after 13:27 IhrFussel https://pastebin.com/DD8wP0Pd 13:29 IhrFussel Keep in mind that ALL 'before' values were crafted with 1 item 13:29 IhrFussel While most 'after' values executed 5-10 items 13:30 BuckarooBanzai Oh, that does not look good: i have huuge lag spikes in v5.0 with 1 player logged in :( https://test.pandorabox.io/grafana/d/cACE6ppik/overview?refresh=5s&orgId=1 13:30 BuckarooBanzai A normal graph in comparison: https://pandorabox.io/grafana/d/cACE6ppik/overview?refresh=5s&orgId=1 13:31 IhrFussel BuckarooBanzai, profiler.load = true in conf 13:32 BuckarooBanzai IhrFussel: already investigating... maybe some db- or hd-issue, it runs on an old server :/ 13:33 IhrFussel Check top first 13:33 IhrFussel Check iowait and st values 13:33 IhrFussel (st will only be relevant if you server is a VPS/shares resources with other customers) 13:34 rubenwardy interesting how there's such a mixed response to it - some people say 5.0 improves performance a lot, others say the opposite 13:34 rubenwardy did you migrate to the auth.sql db? 13:34 rubenwardy what lag is that? Minetest m_lag? 13:34 BuckarooBanzai rubenwardy: not yet, i have no login-traffic on that server yet... 13:35 rubenwardy well, that crosses that out :) 13:35 rubenwardy but yeah, profiler is a good idea 13:35 BuckarooBanzai rubenwardy: it's a self-rolled value and from /status: https://github.com/thomasrudin-mt/monitoring/blob/master/builtin/lag.lua 13:35 rubenwardy and also checking that thing which tells you when files are modified, to see if it matches a save 13:36 rubenwardy I did this on my server and found some issues ^ 13:37 BuckarooBanzai the "thing" ? 13:38 rubenwardy inotifywait -me modify,create,delete -r . 13:38 rubenwardy in .minetest/worlds/world 13:38 BuckarooBanzai ah :) 13:38 rubenwardy it'll print when files are modified, created, or deleted 14:38 IhrFussel Do ABMS run per player or doesn't matter how many players are in one mapblock it will only run once per interval? 14:43 rubenwardy player independent 14:43 rubenwardy only dependent on active blocks 18:28 TenPlus1 hi folks :) 18:28 TenPlus1 hey Krock 18:28 Krock hi TenPlus1 18:29 TenPlus1 I see 5.0 is now in stable... yay :) 18:29 TenPlus1 will android release be coming soon ? 18:29 rubenwardy already has 18:29 rubenwardy nerzhul is keeping it in beta until 5.0.1 18:30 TenPlus1 oh, so players can go into google play or f-droid and install 5.0 ? 18:30 rubenwardy yes 18:30 TenPlus1 w00t! will look into upgrading server soon 18:31 nerzhul yep i should look for returns on 5.0.0 as we are not releasing a 5.0.1 i don't know why 18:31 TenPlus1 hi nerzhul 18:32 rubenwardy https://github.com/minetest/minetest/milestone/14 18:32 rubenwardy nerzhul /6 18:32 rubenwardy nerzhul ^ 18:32 nerzhul hi TenPLus1 18:32 nerzhul if we are too late just release the current blocker fixes, we can solve others later, as i already said 18:32 rubenwardy 5.0.1 should probably be seen as blockers v2 18:32 nerzhul 1 week since latest release 18:32 rubenwardy yeah 18:33 * TenPlus1 is eager to test 18:34 benrob0329 o/ TenPlus1 18:35 TenPlus1 hi benrob :) 18:36 TenPlus1 saw the mt5.0 showcase vid that you assisted nathan with :) was hoping he'd show more tho :P 18:37 TenPlus1 ooh nice 18:39 Krock amazing resolution, benrob0329. I cannot watch the beauty of Minetest in realtime 18:39 TenPlus1 watching "illumination of Illuna" :D 18:41 benrob0329 Krock: heh, it happens with 1080p60 video :P 18:41 TenPlus1 quality is indeed amazing :) kewl vids dude 18:43 benrob0329 Thank you :D 18:46 TenPlus1 le click 18:47 Krock le like 18:48 TenPlus1 heh funny :) 18:48 * TenPlus1 secretly smashes the 5 button on phone 19:05 TenPlus1 wut ??? 19:05 TenPlus1 lol 19:26 TenPlus1 laters all :) 19:28 Krock ditto. see ya next weekend 19:39 Fixer wayback machine runs on dial up or smth... 19:50 benrob0329 Hello Megaf 21:30 benrob0329 nerzhul: how long does it usually take for the FreeBSD minetest port to update? 21:42 Fixer wtfuuuck, pinterest now spams general google search? fuck this shit 21:45 red-045 like more then the usual? 21:45 red-045 pinterest seems to show up a lot if google can't find any good results for something 21:46 Fixer probably 21:46 Fixer googling some rare stuff 22:16 nerzhul benrob i don't know 22:18 VanessaE in all the years I've been on the web and thousands of searches, not once has a pinterest result ended up being the thing I want. 22:19 VanessaE (and rarely has it even been particularly relevant)