| *** YukariYakumo <YukariYakumo!~Yukari@user/yukari> has joined #bzflag | 00:56 | |
| TimRiker | I've not decided what I'm working on next, but I hope to start isolating code. Create a routine to display addresses, and change everyplace that we convert network addresses to text to all the same routine. Things like that. | 01:40 |
|---|---|---|
| TimRiker | blast007, if you'd check in your experiments someplace, I would take some time to look them over. | 01:41 |
| TimRiker | bzflag and bzadmin are working against the ipv6 enabled my-dev.bzflag.org which seems like a step forward. | 01:51 |
| TimRiker | the new site is also http2 enabled. | 01:52 |
| TimRiker | s/new/dev/ | 01:52 |
| BZNotify | feature_nameport @ bzflag-bzfls: timriker pushed 1 commit (https://github.com/BZFlag-Dev/bzflag-bzfls/compare/e07d0b478704...a22eff2d4f45): | 01:57 |
| BZNotify | feature_nameport @ bzflag-bzfls: timriker a22eff: add key to form (https://github.com/BZFlag-Dev/bzflag-bzfls/commit/a22eff2d4f4551bbd156c06afa004e0a6215dfb3) | 01:57 |
| BZNotify | bzflag-bzfls: timriker synchronized pull request #1 "Token nameport support" (https://github.com/BZFlag-Dev/bzflag-bzfls/pull/1) | 01:57 |
| BZNotify | feature_nameport @ bzflag-bzfls: timriker pushed 1 commit (https://github.com/BZFlag-Dev/bzflag-bzfls/compare/a22eff2d4f45...240edcccf710): | 02:03 |
| BZNotify | feature_nameport @ bzflag-bzfls: timriker 240edc: no REGISTER or CONFIRM, yet? (https://github.com/BZFlag-Dev/bzflag-bzfls/commit/240edcccf7105c5715858f7b331fec51de3853ac) | 02:03 |
| BZNotify | bzflag-bzfls: timriker synchronized pull request #1 "Token nameport support" (https://github.com/BZFlag-Dev/bzflag-bzfls/pull/1) | 02:03 |
| BZNotify | feature_nameport @ bzflag-bzfls: timriker pushed 1 commit (https://github.com/BZFlag-Dev/bzflag-bzfls/compare/240edcccf710...9c11b3fff84d): | 02:04 |
| BZNotify | feature_nameport @ bzflag-bzfls: timriker 9c11b3: no REGISTER or CONFIRM, yet? (https://github.com/BZFlag-Dev/bzflag-bzfls/commit/9c11b3fff84dfe3853d197670a031e6e02b556f1) | 02:04 |
| BZNotify | bzflag-bzfls: timriker synchronized pull request #1 "Token nameport support" (https://github.com/BZFlag-Dev/bzflag-bzfls/pull/1) | 02:04 |
| *** YukariYakumo <YukariYakumo!~Yukari@user/yukari> has quit IRC (Quit: Quit.) | 03:41 | |
| *** Sgeo <Sgeo!~Sgeo@user/sgeo> has quit IRC (Read error: Connection reset by peer) | 04:45 | |
| *** Sgeo <Sgeo!~Sgeo@user/sgeo> has joined #bzflag | 04:47 | |
| *** _I_Died_Once <_I_Died_Once!~I_Died_On@c-73-184-170-223.hsd1.ga.comcast.net> has joined #bzflag | 04:55 | |
| *** I_Died_Once <I_Died_Once!~I_Died_On@c-73-184-170-223.hsd1.ga.comcast.net> has quit IRC (Ping timeout: 255 seconds) | 04:57 | |
| *** _I_Died_Once <_I_Died_Once!~I_Died_On@c-73-184-170-223.hsd1.ga.comcast.net> has quit IRC (Ping timeout: 268 seconds) | 05:05 | |
| TimRiker | useful reading: https://www.kame.net/newsletter/19980604/ | 06:03 |
| *** Cobra_Fast_ is now away: vacant | 06:18 | |
| *** Cobra_Fast_ is back | 06:18 | |
| *** Sgeo <Sgeo!~Sgeo@user/sgeo> has quit IRC (Read error: Connection reset by peer) | 08:09 | |
| *** FastLizard4 is back | 08:24 | |
| *** FastLizard4 is now away: AWAY from keyboard | 08:53 | |
| TimRiker | if our address2string routines converted ::ffff:a.b.c.d back to a.b.c.d for display, could we get by with one ipv6 listener? experimenting with test code here. | 08:58 |
| *** FastLizard4 is now away: GONE - Screen Detached and Disconnected from IRC (I'm probably asleep, at work, or doing something in real life) | 09:24 | |
| blast007 | TimRiker: I guess my thought was that it would be a problem for people that wanted to bind to specific interfaces | 10:32 |
| blast007 | it's probably less common to have multiple IPv4 addresses on a host, but I see my old host XLHost still offers dedicated servers with 5 or 13 IPv4 addresses (with the option to add more, with justification), and a /64 of IPv6 for an extra $2/month | 10:36 |
| blast007 | and yeah, I've been referencing that link a bit, as well as https://beej.us/guide/bgnet/html/ and https://www.rfc-editor.org/rfc/rfc4038 | 10:43 |
| *** _I_Died_Once <_I_Died_Once!~I_Died_On@c-73-184-170-223.hsd1.ga.comcast.net> has joined #bzflag | 12:19 | |
| *** disco <disco!~disco@81.187.95.100> has quit IRC (Quit: Lost terminal) | 15:20 | |
| *** disco <disco!~disco@81.187.95.100> has joined #bzflag | 15:21 | |
| *** Sgeo <Sgeo!~Sgeo@user/sgeo> has joined #bzflag | 15:49 | |
| *** FastLizard4 is back | 16:20 | |
| *** FastLizard4 is now away: AWAY from keyboard | 16:25 | |
| *** Optic_Delusion_ <Optic_Delusion_!~Optic_Del@2600:4041:19d:2d00:e8bc:2748:a171:a980> has joined #bzflag | 19:26 | |
| *** Optic_Delusion <Optic_Delusion!~Optic_Del@pool-173-75-132-196.pitbpa.fios.verizon.net> has quit IRC (Ping timeout: 255 seconds) | 19:27 | |
| TimRiker | playing here: https://github.com/timriker/ipv6 | 21:07 |
| TimRiker | nc -N -u localhost 7002 < test running server gets a Message too long error. I don't recall how we determine max mtu for our udp packets. | 22:09 |
| TimRiker | nc -N -i 1 localhost 7002 < README.md is interesting. or with a -u | 22:36 |
| TimRiker | allowing -i multiple times could be useful for some folks. I'm talking about the default. I think a single default binding on :: should work for most people. I've tested on WSL which does not have IPv6 connectivity. If I bind to 0.0.0.0 then windows processes can see it, but on :: only linux processes see it. a -i 0.0.0.0 should be all the work around needed. | 22:42 |
| TimRiker | on display, do we show :::5154 like netstat? or [::]:5154 like curl, etc expect? more questions... | 22:43 |
| TimRiker | then there's the question of banlists and how we show/match cidr blocks. | 22:55 |
| TimRiker | how/where does master-bans.txt get updated? | 22:57 |
| blast007 | master-bans.txt doesn't really get updated anymore | 23:05 |
| blast007 | the server doesn't reload it, so it was kinda useless in most cases | 23:05 |
| blast007 | one part of my reasoning for wanting to use two sockets (one IPv4 and one IPv6) as the default was so that multiple sockets was a well tested code path | 23:07 |
| blast007 | I'm gonna try making a sample server program based around some new code I'm writing and a simplified NetHandler class. | 23:08 |
| blast007 | I'll throw that in a repo or a Gist when I have something | 23:09 |
| TimRiker | blast007++ | 23:19 |
| TimRiker | sockaddr_storage is 128 bytes, where sockaddr_in6 is only 28 bytes. Do you think we'll use anything other than v4 and v6? I tend to think v6 space is enough, but I don't want to see ::ffff:fff: everywhere. If our display removes that, then I think sockaddr_in6 should work for us. | 23:22 |
| TimRiker | not looked to see what network messages contain ips. If we have some, we probably want a prefix for AF family followed by the binary network order ip bytes. | 23:23 |
| blast007 | there's a comment here that sockaddr_storage is "64-bit aligned for efficiency" https://stackoverflow.com/questions/15075748/sockaddr-storage-size-of-128-bytes | 23:47 |
| blast007 | main message we have that contains an IP is MsgAdminInfo and I think our ping packet? | 23:47 |
| blast007 | If we can improve the server administrator UI, we could probably get rid of MsgAdminInfo | 23:50 |
| blast007 | it's mainly there right now because there's no server-side history of players, so to ban someone that left you have to type the the IP you see in the chat backlog | 23:50 |
| TimRiker | in your multiple listening plan, how do we know which source ip to use? will playerinfo have a server sockaddr as well as a player sockaddr? | 23:56 |
| TimRiker | the third comment there is the big one. sockaddr_un stores a path. I doubt we'll ever use that. | 23:58 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!