08:46 <+bridge> [ddnet] omg new devs 10:53 <+bridge> [ddnet] I should maybe look into reiserfs for storing teehistorian files, the median file size is only 18kb 10:56 <+bridge> [ddnet] Why? Reiserfs seems really outdated. See for example https://www.phoronix.com/scan.php?page=article&item=reiser4-linux-417&num=3 10:56 <+bridge> [ddnet] > To little surprise given that Reiser4 doesn't see too much optimization/feature work these days, Reiser4 tended to struggle compared to today's EXT4, F2FS, Btrfs, and XFS competition that are within the mainline Linux kernel. In fact, it's already been a decade since Hans Reiser was last able to contribute to the file-system's development and since then there has just been Edward Shishkin and few other contributors. Given the small bus count 10:56 <+bridge> [ddnet] I've never had problems with ext4. 10:59 <+bridge> [ddnet] hm, my lookups have been taking very long lately, and I'd heard after a couple million files inside a folder ext4 starts to have some trouble 11:01 <+bridge> [ddnet] then let's not put millions of files into a single folder but presort by prefix 11:01 <+bridge> [ddnet] and create subdirectories for those prefixes 11:04 <+bridge> [ddnet] That's a good idea actually 11:04 <+bridge> [ddnet] I love how my mind went to some weird filesystem no one used in a decade instead of that :D 11:08 <+bridge> [ddnet] Lookups on my teehistorian server or yours? 11:15 <+bridge> [ddnet] mine 11:23 <+bridge> [ddnet] Hm, I guess what's slow is enumerating, not indexing 11:28 <+bridge> [ddnet] u 11:58 <+bridge> [ddnet] pro-tip, having a couple instances of rsync running in parallel better saturates your connection if your files are very small 12:25 <+bridge> [ddnet] @Learath2 as in it uses all ur net speed? 12:25 <+bridge> [ddnet] @Learath2 answered on the pull request, not sure what happens there for what reason sometimes. Would be happy if we could get it in, the debug messages of the bot are still kinda failing since check_ddnet apparently debugs some of the stuff to stderr. what I find crazy is that it was never a problem on my test server which I can honestly not explain at all. so one of the commits fixes it and if you want to update the twmap binaries they will 12:26 <+bridge> [ddnet] i had a router handed to me by my isp when they installed the internet, it couldnt handle 200 connections at the same time and internet got down for some time (i was torrenting animu), i bought a more decent router and it works flawlessly 12:26 <+bridge> [ddnet] stupid isps 12:27 <+bridge> [ddnet] and i can ssh into my router 12:27 <+bridge> [ddnet] altho i have to tell ssh to use some ancient cyphers 12:27 <+bridge> [ddnet] KexAlgorithms +diffie-hellman-group1-sha1Ciphers 3des-cbc 12:27 <+bridge> [ddnet] KexAlgorithms +diffie-hellman-group1-sha1 12:27 <+bridge> [ddnet] Ciphers 3des-cbc 12:27 <+bridge> [ddnet] this 12:38 <+bridge> [ddnet] @Ryozuki as in it uses more of it atleast 12:39 <+bridge> [ddnet] It was doing 200K down, with 8 jobs running it does 2.6M down 12:41 <+bridge> [ddnet] @Patiga how about I do both? πŸ˜„ 12:42 <+bridge> [ddnet] hm, seems you didn't answer on the PR? deleted your answer? 12:44 <+bridge> [ddnet] GNU Parallel is also very nice to use, if you ever want to parallelise some tasks 12:48 <+bridge> [ddnet] hm it has the 'Pending' tag, not sure what that means 12:48 <+bridge> [ddnet] > Good point, I would agree by looking at the code, but testing it in my test server shows that I can't use it in #submit-maps anyways. Since only testers can use the command, I don't think we need to investigate it much further 12:49 <+bridge> [ddnet] @Patiga You accidentally started a review, you need to click send single comment πŸ˜› 12:50 <+bridge> [ddnet] hm how do I get out of the review? o.o 12:51 <+bridge> [ddnet] You just delete your review comment and resend it as a single 12:51 <+bridge> [ddnet] or you can end the review on the top right 12:52 <+bridge> [ddnet] okay I feel blind, I only see `Submit review` 12:52 <+bridge> [ddnet] anyway, did you try using it on a submission that you already accepted? that's what it looked like it'd work on for me 12:53 <+bridge> [ddnet] so using the command $edit in the #πŸ“¬submit-maps channel on a map that got accepted? 12:54 <+bridge> [ddnet] You can submit it as a comment, the top option. Or you can cancel your review by deleting the pending comment 12:54 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/869170702459240468/unknown.png 12:54 <+bridge> [ddnet] yep 12:54 <+bridge> [ddnet] That is accepted as in already has a channel 12:56 <+bridge> [ddnet] seems like it doesnt do anything, wanna get on my test server? 12:57 <+bridge> [ddnet] sure 13:13 <+bridge> [ddnet] @Patiga btw you don't have to put any comment in the overall review 13:18 <+bridge> [ddnet] https://github.com/ddnet/ddnet-discordbot/pull/23#pullrequestreview-714708112 this one is at the correct place though, no? 13:34 <+bridge> [ddnet] http headers are weird \:D 13:35 <+bridge> [ddnet] can they be trusted in any way? I mean i can set whatever i want as origin and forwarded for etc 13:50 <+bridge> [ddnet] Ones inserted by your webserver can be trusted if your webserver says so, the others are not to be trusted 13:55 <+bridge> [ddnet] You could have left that empty and it wouldn't have sent a comment at all 14:08 <+bridge> [ddnet] https://gchq.github.io/CyberChef/ very useful tool 16:06 <+bridge> [ddnet] France perma is laggy 16:07 <+bridge> [ddnet] Fix this please 16:07 <+bridge> [ddnet] I don't have any issues on fra 16:29 <+bridge> [ddnet] :ddnet:Suggestion :ddnet: 16:29 <+bridge> [ddnet] What if there was a tick box where if you have more than 1 game skin or emoticon pack every tee except yours is assigned a random one to keep? 16:29 <+bridge> [ddnet] Like, everyone on a team has a different hammer look or grenade but they all function the same. 16:42 <+bridge> [ddnet] i think thats actually funny, but its probably not going to be implemented. 17:23 <+bridge> [ddnet] yea thats probably too hard to make for just a funny feature 18:11 <+bridge> [ddnet] https://asia.nikkei.com/Business/Tech/Semiconductors/TSMC-eyes-Germany-as-possible-location-for-first-Europe-chip-plant 18:11 <+bridge> [ddnet] :poggers: 18:32 <+bridge> [ddnet] why does server reallocates character on tryrespawn? 18:32 <+bridge> [ddnet] why does server reallocate character on tryrespawn? 18:36 <+bridge> [ddnet] ig each time you die your character is destroyed (which is logic), so when you spawn back it does reallocate 18:36 <+bridge> [ddnet] cplayer should be the player itself, ccharacter the tee ingame 18:40 <+bridge> [ddnet] but is there any need to reallocate it since destroy() makes character isalive = false 19:02 <+bridge> [ddnet] you could check git blame to find out why 19:02 <+bridge> [ddnet] I don't know a good reason off the top of my head 19:12 <+bridge> [ddnet] The character is recreated but characters are created inside a pool, so they are always in the same place. So no actual allocation happens 19:13 <+bridge> [ddnet] Did you not wonder why the odd `new(m_ClientID) CCharacter(...)` and not just `new CCharacter(...)`? πŸ˜„ 19:30 <+bridge> [ddnet] i knew about it but anyway why does it happen? xd 19:35 <+bridge> [ddnet] oh wrong question 19:36 <+bridge> [ddnet] why *should* it happen? 19:36 <+bridge> [ddnet] why not 19:38 <+bridge> [ddnet] also noticed that we store m_Tile(F)Index in CCharacter but we can just use it in local scopes 19:38 <+bridge> [ddnet] also noticed that we store m_Tile(F)Index in CCharacter but we can just use them in local scopes 20:12 <+bridge> [ddnet] Anyone has experience with getting filtered by the provider? (Home Connection)Well tbh I am not sure who is filtering me. But I can not reach half the internet including my own servers.All started when I did a stress test of my webapp I sent maybe too many requests \:justatest\: 20:13 <+bridge> [ddnet] :justatest: 20:16 <+bridge> [ddnet] hmm idk 20:16 <+bridge> [ddnet] first step is to not use ur isp dns 20:16 <+bridge> [ddnet] http over dns helps too 20:16 <+bridge> [ddnet] dns over http 20:16 <+bridge> [ddnet] * 20:16 <+bridge> [ddnet] firefox has the option 20:17 <+bridge> [ddnet] i dont think chrome does cuz google 20:17 <+bridge> [ddnet] oh they do 20:30 <+bridge> [ddnet] I can not connect to my vps via ip either 20:30 <+bridge> [ddnet] Only things that work that I tested are Netflix YouTube tw master and tw servers