Timestamps are in GMT/BST.
| 00:38:34 | | raptor has joined |
| 00:38:34 | | ChanServ sets mode +o raptor |
| 00:49:13 | | Zoomber has joined |
| 00:49:13 | | ChanServ sets mode +v Zoomber |
| 00:49:21 | Zoomber | hey raptor |
| 00:49:23 | Zoomber | sam686 |
| 01:26:32 | raptor | hi Zoomber |
| 01:29:45 | Zoomber | hey raptor |
| 01:29:52 | raptor | hi |
| 01:32:27 | sam686 | hi |
| 01:53:04 | | raptor Quit (Ping timeout: 245 seconds) |
| 02:00:23 | | raptor has joined |
| 02:00:23 | | ChanServ sets mode +o raptor |
| 03:44:50 | sam686 | now i starting to here complaints that afk players in a server, even though they don't respawn, are included in game stats... |
| 03:45:20 | raptor | not sure what to do about that... |
| 03:46:06 | raptor | maybe have a flag that says they played at least 25% of the match or something? |
| 05:09:39 | raptor | hi sam686, you around? |
| 05:25:00 | sam686 | now i am.. |
| 05:27:13 | raptor | old 015a aren't getting update messages, i think - MasterServerConnection::readConnectRequest has changed - 015a clients are detected as servers |
| 05:27:52 | sam686 | the old 015a uses the old master |
| 05:28:35 | raptor | ah that's right |
| 05:28:40 | raptor | so i don't need to worry.. |
| 05:30:12 | sam686 | the old 015a master appears to be running, but with a bug that seems to keep a bunch of servers / players listed |
| 05:30:51 | sam686 | have fun reading this old master: http://eykamp.com/bitfighter/bitfighterStatus.json |
| 05:31:10 | raptor | oh wow, it's growing |
| 05:32:26 | sam686 | the old master appears to have a bug in TNL, that when milliseconds overflows, it doesn't auto-disconnect the connections to client / server |
| 05:32:57 | sam686 | i think i might have fixed it for 017 in TNL.. |
| 05:33:01 | raptor | you fixed that right? |
| 05:33:05 | raptor | i think i remember you did.. |
| 05:34:00 | sam686 | at least the old 015a master displays outdated version.. |
| 05:53:59 | | BFLogBot - Commit 42196fcd6680 | Author: buckyballreaction | Log: Update master server to use an INI file for its settings (master.ini) |
| 05:54:46 | raptor | man that was work |
| 05:58:56 | raptor | sam686: maybe we should auto kick from the server if someone has been idle for a whole game? |
| 05:59:19 | sam686 | maybe, not sure... |
| 06:03:41 | Zoomber | when will karamazovapy get back on here |
| 06:03:53 | sam686 | on another side, i am trying to start programming my own 3D game, partially from stratch. http://sam686.maxhushahn.com/upload/my_cpp_program.PNG i have some process, though.. |
| 06:04:03 | Zoomber | as raptor works viciously on his coding, I work viciously on my synth-ing |
| 06:04:23 | raptor | sam686: neat! |
| 06:04:51 | sam686 | i haven't seen karamazovapy on the forum (his forum last visited date is a few days ago) |
| 06:07:46 | raptor | uh oh, past my curfew |
| 06:07:50 | raptor | good night all! |
| 06:08:40 | | raptor Quit () |
| 06:09:31 | Zoomber | hey sam686, did you ever try working on that animated text rotater in game? |
| 06:12:17 | sam686 | i haven't... |
| 06:13:03 | sam686 | i was spending most of my time, trying to make my new 3D game or program... |
| 06:16:01 | Zoomber | can you still give it a shot? |
| 06:18:16 | sam686 | maybe, but probably not tonight... |
| 06:19:45 | Zoomber | do you just not want to do it?, I can try to do it myself, I'm just really busy right now working with my music, and don't really have the time to, nonetheless probably the skill to get past whatever problem i encounter when trying to do it. |
| 06:20:09 | Zoomber | I feel like it would be as easy as copying whatever code there is for speedzones into text-items, but i couldn't know for sure |
| 06:22:06 | sam686 | i probably don't want do do it... as i am thinking what good does it do if it doesn't affect gameplay? |
| 06:22:39 | Zoomber | what if we did it with barriers? |
| 06:23:26 | sam686 | i am not sure if it is possible to move barriers in the middle of game, due to the way the barrier data is send to client |
| 06:23:45 | Zoomber | its sent differently than a speedzone is? |
| 06:23:48 | sam686 | if barriers did get moved (like in /edit mode), it fails to update to client |
| 06:24:01 | Zoomber | you cant pivot a barrier? |
| 06:24:06 | Zoomber | like, circular one point? |
| 06:24:54 | sam686 | barrier is send through some S2c commands, making updading in mid game difficult |
| 06:25:26 | sam686 | most other opjects uses Ghostable system which allows setBitMask / packUpdate to work in mid game |
| 06:25:46 | Zoomber | ahh |
| 06:26:48 | sam686 | TextItem and LineItem happen to be in the ghostable system, allowing easy update to position of text . lines |
| 06:28:24 | sam686 | a second problem with barriers is how a 20 point barrier (not polygon barrier) get split into 19 individual 2-point barriers while server loading level. |
| 06:35:17 | sam686 | i need to get to bed, bye.. |
| 06:41:24 | | sam686 Quit (Ping timeout: 245 seconds) |
| 08:15:16 | | Zoomber Quit (Quit: Zoomber) |
| 13:33:50 | | watusimoto has joined |
| 13:33:52 | | ChanServ sets mode +o watusimoto |
| 14:14:44 | | raptor has joined |
| 14:14:44 | | ChanServ sets mode +o raptor |
| 14:58:58 | watusimoto | !bug |
| 14:58:58 | BFLogBot | To enter a bug, please make sure it is reproducible and then go to http://code.google.com/p/bitfighter/issues/list | Also, see current running bug list: http://bitfighter.org/wiki/index.php/Running_Bug_List |
| 15:02:14 | raptor | hi |
| 16:23:48 | watusimoto | hi |
| 16:23:55 | raptor | hi |
| 16:25:38 | watusimoto | hi |
| 16:26:34 | raptor | master.ini should help in the long run... |
| 16:26:50 | watusimoto | yeah, I was surprised to see you did that |
| 16:26:55 | raptor | i wasn't quite sure what to do about the motd, but I think I handled most cases of previous use that i could find |
| 16:27:02 | watusimoto | not too hard, I hope? |
| 16:27:16 | raptor | nope, it was actually easier than i thought |
| 16:27:20 | raptor | that INI class is nice |
| 16:27:45 | watusimoto | yes; we've improved it gradually over time |
| 16:28:09 | watusimoto | we should see if we can push the results upstream; not sure to what extent we infected it with S32 and the like |
| 16:28:34 | watusimoto | I know I renamed a bunch of methods |
| 16:28:42 | raptor | I've personally refactored 80% of that class |
| 16:28:45 | watusimoto | and did a lot of formatting |
| 16:28:53 | watusimoto | well, ok then |
| 16:29:08 | watusimoto | maybe we can spin it off as a fork of the original |
| 16:29:13 | watusimoto | or maybe we can not worry about it |
| 16:29:13 | raptor | because all the internal variables did make sense... |
| 16:29:22 | raptor | keys were values, and sections were keys |
| 16:29:25 | raptor | etc |
| 16:29:34 | watusimoto | right |
| 16:29:41 | watusimoto | that was very confusing |
| 16:29:53 | raptor | not any more! |
| 16:30:00 | watusimoto | :-) |
| 16:30:27 | watusimoto | so I found that there are only 2^10 (1024) different keys for stringTableEntries |
| 16:30:37 | watusimoto | not sure what happens when those are exhausted |
| 16:30:59 | raptor | interesting |
| 16:33:47 | watusimoto | you can store more, but transmission is limited to 10 bits |
| 16:36:05 | watusimoto | looks like it uses some variant of reusing indexes, but not on a most recent used basis |
| 16:36:29 | watusimoto | or maybe... tere's a lot of complex logic going on |
| 16:36:40 | watusimoto | ConnectionStringTable::writeStringTableEntry if you care |
| 16:37:07 | raptor | sometimes I feel like this still isn't our code... |
| 16:37:27 | watusimoto | the network stuff certainly isn't; and that's a good thing :-) |
| 16:37:34 | raptor | hehe, yep |
| 16:39:24 | watusimoto | I'll try to get saving of level stats working tonight |
| 16:39:37 | raptor | that's all that's really left, right? |
| 16:39:55 | raptor | (for 017) |
| 16:43:29 | watusimoto | level stats, achievement working, playerStats insertion (which is really easy) |
| 16:43:42 | raptor | oh yeah, the achievement |
| 16:43:47 | watusimoto | achievement is mostly working, but untested |
| 16:43:52 | raptor | do you know how you want it to be displayed? |
| 16:43:55 | watusimoto | I *think* all the protocols are in place |
| 16:44:01 | watusimoto | display on scoreboard for now |
| 16:44:05 | watusimoto | next to player's name |
| 16:44:16 | watusimoto | and at some point in forums |
| 16:45:33 | watusimoto | looks like achemevent saving is in place on master |
| 16:45:42 | watusimoto | just need to call s2mAcheivementAchieved() under the right circumstances |
| 16:46:02 | watusimoto | maybe reading from the database is still needed |
| 17:31:49 | | watusimoto Quit (Ping timeout: 260 seconds) |
| 17:47:14 | | Watusimoto has joined |
| 18:17:02 | | sam686 has joined |
| 18:17:02 | | ChanServ sets mode +v sam686 |
| 18:18:17 | raptor | are all three of us actually here at the same time? |
| 18:25:26 | | BFLogBot - Commit e4b959fca5bb | Author: watusim...@bitfighter.org | Log: Level stats SHOULD make it to master -- need to rebuild master to test |
| 18:25:34 | raptor | want me to rebuild? |
| 18:26:53 | Watusimoto | sure -- I tried doing it myself, but things got messy |
| 18:27:41 | Watusimoto | I did hg pull and got abort: repository default not found! |
| 18:28:19 | raptor | did you do it in the 'bitfighter' directory? |
| 18:28:32 | raptor | compiling |
| 18:28:57 | Watusimoto | yes |
| 18:29:23 | Watusimoto | got yr msg |
| 18:29:29 | raptor | really!>? |
| 18:29:54 | Watusimoto | yes |
| 18:30:03 | raptor | great! |
| 18:30:07 | raptor | i've always wanted to do that... |
| 18:30:29 | Watusimoto | :-) |
| 18:31:13 | raptor | compilied |
| 18:31:16 | raptor | compiled |
| 18:31:20 | raptor | and restarted |
| 18:31:58 | Watusimoto | testing |
| 18:32:29 | Watusimoto | CRASH |
| 18:32:49 | raptor | master is still up |
| 18:37:29 | Watusimoto | problem is something on server (not master) |
| 18:37:37 | Watusimoto | hasn't even tried to communicate yet |
| 18:37:54 | Watusimoto | server game's getConnectionToMaster(); is coming back NULL |
| 18:39:03 | Watusimoto | stats may have been sent... |
| 18:39:29 | Watusimoto | checking db... |
| 18:40:10 | Watusimoto | seems to work |
| 18:40:37 | raptor | potato |
| 18:41:12 | Watusimoto | ok, I see the problem |
| 18:41:21 | Watusimoto | it takes a while for server to connect to master |
| 18:41:33 | Watusimoto | but it wants to send stats the moment the level loads |
| 18:54:21 | Watusimoto | do we want to keep a count of how often we see a level in level stats? |
| 18:54:52 | raptor | explain |
| 18:55:10 | raptor | can't we just determin that from a query? |
| 18:55:28 | Watusimoto | I don't write duplicate entries to the db |
| 18:55:50 | raptor | but a game is tied to a level, right? |
| 18:55:58 | Watusimoto | server doesn't send repeated data, and master checks db to avoid dupe levels |
| 18:56:04 | raptor | so we can count the games that point to that level? |
| 18:56:07 | Watusimoto | I don't think it is yet |
| 18:56:11 | Watusimoto | but will be, yes |
| 18:56:19 | Watusimoto | oh, that's true |
| 18:56:27 | | LordDVG has joined |
| 18:56:39 | raptor | ok, if the game -> level, then we're good, i think |
| 18:56:47 | Watusimoto | that will tell us completed games with that level, which I suppose is a reasonabble metric |
| 18:57:00 | Watusimoto | right now, only level name is stored with game |
| 18:57:13 | Watusimoto | at some (near) future point, we'll need to change that to level_id |
| 18:57:18 | raptor | that should probably change to id or hash |
| 18:57:21 | Watusimoto | but we need some level data first |
| 18:57:27 | raptor | ah yes |
| 18:57:29 | Watusimoto | to join up the older data |
| 18:57:30 | raptor | that's right |
| 18:57:36 | raptor | we discussed this, huh... |
| 18:57:40 | raptor | we'd have to wait |
| 18:57:45 | Watusimoto | but yes, we'll get a freq count that way, so that works |
| 18:58:00 | Watusimoto | I was thinking of a scheme where everytime a level gets loaded we'd count it |
| 18:58:07 | Watusimoto | but there is no real point to that |
| 18:58:19 | Watusimoto | we do grab level info when a level is loaded |
| 18:58:26 | Watusimoto | to try to build the database faster |
| 18:58:37 | raptor | yeah, because the same level is loaded a million times everytime i host |
| 18:58:39 | Watusimoto | but there will be levels without games pointing to them |
| 18:58:49 | raptor | and i always switch to antoher |
| 18:58:54 | Watusimoto | right -- don't want a million entries in the level tab;e |
| 19:05:55 | Watusimoto | are we saving gameType as int rather than string? |
| 19:06:05 | Watusimoto | I think it should be string |
| 19:06:11 | Watusimoto | (in level stats file) |
| 19:06:47 | raptor | there is some enum somewhere that it maps to |
| 19:06:56 | raptor | always been like that i think.. |
| 19:07:30 | raptor | i think we save as string somewher ein the database |
| 19:08:07 | raptor | stats_game.game_type |
| 19:09:26 | Watusimoto | I changed it to string; maybe I should change it back to int... we send it to server as string |
| 19:09:41 | Watusimoto | what's the downside to storing the string? |
| 19:09:47 | Watusimoto | more disk space? |
| 19:09:57 | sam686 | string take up more space in mem |
| 19:10:12 | raptor | a little, but it would make sense to store as int if it was an ID to another table with the strings in it |
| 19:10:28 | Watusimoto | how do we go from the string we have on the master to the id we write in the table? |
| 19:10:36 | Watusimoto | i.e. can we do it in one query? |
| 19:10:36 | raptor | stats_game_type |
| 19:10:53 | raptor | uhh |
| 19:10:57 | raptor | which query? |
| 19:11:23 | Watusimoto | tell you what |
| 19:11:31 | Watusimoto | one of you guys can rewrite this method: |
| 19:11:31 | Watusimoto | DatabaseWriter::insertLevelInfo |
| 19:11:38 | Watusimoto | to write the correct int |
| 19:12:19 | Watusimoto | change the table back to an int column; I'll delete all the records I've written in there so the table will be clean |
| 19:12:26 | Watusimoto | acutally, I'll just change the column to int |
| 19:12:32 | Watusimoto | as I have the page open |
| 19:13:09 | raptor | what table are you editing? |
| 19:13:36 | raptor | we save game_type in stats_game |
| 19:13:42 | raptor | maybe we shouldn't? |
| 19:14:05 | Watusimoto | ok |
| 19:14:16 | Watusimoto | stats_level |
| 19:14:27 | Watusimoto | I just changed the game_type from string to int |
| 19:14:32 | Watusimoto | I can change it back to string |
| 19:14:36 | Watusimoto | and everything will work |
| 19:14:40 | Watusimoto | or we can leave it as int |
| 19:14:53 | raptor | ah.. stats_level |
| 19:14:54 | Watusimoto | and change DatabaseWriter::insertLevelInfo to make that write an int (currently it tries to write a string) |
| 19:15:00 | Watusimoto | those are our two choices |
| 19:15:06 | raptor | so we have duplicate data in the DB |
| 19:15:16 | raptor | becasue we keep game_type in stats_game |
| 19:15:21 | raptor | as a string |
| 19:15:26 | Watusimoto | maybe we can remove that eventually |
| 19:15:33 | Watusimoto | when we get things hooked together |
| 19:15:35 | raptor | eventually remove and replace with level id |
| 19:15:37 | raptor | ? |
| 19:15:39 | raptor | yes |
| 19:15:40 | raptor | ok |
| 19:16:06 | Watusimoto | if we go with the int in stats_level, someone else will need to figure out how to get the int from the string |
| 19:16:15 | Watusimoto | because I am way too fried to figure that out |
| 19:16:36 | Watusimoto | though I'm sure it's easy, SQL is NEVER easy for me |
| 19:16:56 | Watusimoto | if we go with string, I can change the db back to string, and this case can be closed |
| 19:17:15 | Watusimoto | if we go with int, we can close the case after the table gets written properly |
| 19:17:21 | Watusimoto | so there are our options |
| 19:17:25 | raptor | we'd need another table that has id -> game_type name |
| 19:17:39 | Watusimoto | I have about 10 minutes to participate in this decision |
| 19:17:44 | raptor | go with whatever easiest for now |
| 19:17:52 | Watusimoto | string is easist |
| 19:18:00 | Watusimoto | I can change the table now -- I have the window open |
| 19:18:01 | raptor | because we need loads of level data before we can properly re-normalize again |
| 19:18:29 | Watusimoto | yes, but that is an unrelated issue |
| 19:18:44 | Watusimoto | ok -- anyone object to storing gameType as string? |
| 19:18:47 | raptor | nah |
| 19:18:51 | Watusimoto | going once... |
| 19:18:54 | | sam686 Quit (Ping timeout: 245 seconds) |
| 19:18:57 | raptor | haha |
| 19:19:15 | Watusimoto | I should give him a second to reconnect |
| 19:20:17 | Watusimoto | sam? |
| 19:20:25 | | sam686 has joined |
| 19:20:25 | | ChanServ sets mode +v sam686 |
| 19:20:33 | | BFLogBot - Commit eff95dd821be | Author: watusim...@bitfighter.org | Log: Address connection to master issues when sending level stats -- now seems to work propery in all tested cases |
| 19:20:36 | Watusimoto | hi sam686 |
| 19:20:36 | raptor | advanced SQL should only be done in the morning with a focused mind |
| 19:20:39 | Watusimoto | going twice |
| 19:20:58 | Watusimoto | sold! |
| 19:21:29 | raptor | i was thinking the two issues were related, because in my head i figure we should do all clean-up at once |
| 19:21:57 | raptor | both stats_game -> level, and normalizing the gametype data |
| 19:23:06 | Watusimoto | well, we still can |
| 19:23:16 | Watusimoto | none of this is irreversible |
| 19:25:12 | Watusimoto | !bug |
| 19:25:12 | BFLogBot | To enter a bug, please make sure it is reproducible and then go to http://code.google.com/p/bitfighter/issues/list | Also, see current running bug list: http://bitfighter.org/wiki/index.php/Running_Bug_List |
| 19:25:36 | | BFLogBot - Commit 560c0193b378 | Author: watusim...@bitfighter.org | Log: Update table schema for sqlite |
| 19:27:44 | Watusimoto | so if a player is the only player in a game, and they are away from their machine, such that the respawn message is up, should the game automatically suspend itself? |
| 19:28:59 | raptor | sam686 brought up the issue last night that some players are just idling happily in several games in a row - and their count goes up in the high scores |
| 19:29:16 | Watusimoto | interesting |
| 19:29:24 | Watusimoto | a different issue |
| 19:29:45 | raptor | also, when a new level starts (and this happened in the BBB) idling players are not taken into account when bulding teams |
| 19:29:58 | raptor | there was one game where it was 5 on 5, and my team had 3 idlers |
| 19:30:02 | raptor | the other, none |
| 19:30:10 | Watusimoto | ah, idlers are added to teams, you mean? |
| 19:30:15 | raptor | yes |
| 19:30:27 | raptor | i bringing up the issue of idlers as a whole |
| 19:30:36 | Watusimoto | so what do we do about that? |
| 19:30:38 | raptor | and the many problems... |
| 19:30:52 | raptor | i was thinking that if you're idling when cycle level happens, you're kicked |
| 19:30:57 | raptor | as one solution |
| 19:30:59 | Watusimoto | we could add idlers to teams last, so at least at the beginning, the number of players was equal |
| 19:31:15 | Watusimoto | or maybe if you idle an entire game |
| 19:31:30 | Watusimoto | a kind of escalation: idle -> kick |
| 19:31:37 | raptor | yes |
| 19:31:54 | raptor | then we'd have to keep track of how long you are idling |
| 19:31:54 | Watusimoto | because you could legit'ly idle at the end of a game |
| 19:32:00 | raptor | agreed |
| 19:32:07 | Watusimoto | just set a flag on client info |
| 19:32:14 | raptor | my idea is very aggresive |
| 19:32:18 | Watusimoto | set it to false at beginning of game |
| 19:32:27 | Watusimoto | set it to true as soon as player is active |
| 19:32:36 | Watusimoto | at end of level, kick everyone with flag == false |
| 19:32:59 | raptor | so a one way switch for the level |
| 19:33:14 | Watusimoto | but what happens if person hosting is idle? |
| 19:33:18 | Watusimoto | does the game end? |
| 19:33:53 | raptor | that's interesting because i've been on a server where that happened - someone was showing off some levels and told everyone they'd be back after a while |
| 19:34:00 | raptor | we played several levels with them idling |
| 19:34:07 | raptor | maybe exclude hoster |
| 19:34:17 | Watusimoto | maybe you shouldn't get kicked, but should go into a nether state |
| 19:34:25 | raptor | observer! |
| 19:34:30 | Watusimoto | how about this |
| 19:34:44 | Watusimoto | at beginning of level, idlers aer not added to teams, and are not counted as players |
| 19:35:01 | Watusimoto | as idlers unidle, then get added to weaker team, just like they joined |
| 19:35:23 | Watusimoto | that fixes sam's observation, and your problem as well |
| 19:35:27 | raptor | that's good |
| 19:35:38 | Watusimoto | so they're connected, but only sort of |
| 19:35:50 | Watusimoto | that leaves my case |
| 19:35:57 | Watusimoto | only one player, who is idle |
| 19:36:01 | Watusimoto | should game suspend? |
| 19:36:10 | Watusimoto | I vote yes |
| 19:36:13 | raptor | sure |
| 19:36:34 | raptor | i somehow missed the 'only one player' part the first time |
| 19:36:35 | Watusimoto | ok then. I'm adding thsoe two cases to the running bug list |
| 19:36:48 | Watusimoto | because they seem like bugs introduced by our new features |
| 19:37:01 | Watusimoto | then I'm working on dinner |
| 19:37:02 | raptor | ok |
| 19:37:07 | raptor | yum |
| 19:38:16 | raptor | it always bothers me when eclipse crashes... |
| 19:38:30 | raptor | it's like my faith in the world has been smashed |
| 19:39:50 | Watusimoto | ok, done |
| 19:39:58 | Watusimoto | but only 4 items on that list, even with the new ones |
| 19:40:08 | Watusimoto | and one is trivial |
| 19:40:46 | Watusimoto | later guys |
| 19:40:51 | raptor | bye |
| 19:45:34 | | Watusimoto Quit (Ping timeout: 260 seconds) |
| 19:57:53 | | zoomber_mbp has joined |
| 20:35:38 | | zoomber_mbp Quit (Quit: zoomber_mbp) |
| 20:37:23 | | LordDVG Quit (Remote host closed the connection) |
| 20:38:53 | | zoomber_mbp has joined |
| 21:00:55 | | Watusimoto has joined |
| 21:05:22 | Watusimoto | ok, who can explain this to me? |
| 21:05:27 | Watusimoto | I run this query on the database |
| 21:05:29 | Watusimoto | SELECT COUNT( * ) |
| 21:05:29 | Watusimoto | FROM `stats_player` |
| 21:05:29 | Watusimoto | WHERE player_name = 'sam686' |
| 21:05:29 | Watusimoto | AND is_authenticated =0 |
| 21:05:33 | Watusimoto | and I get back... |
| 21:05:40 | Watusimoto | 451 |
| 21:05:50 | Watusimoto | I would assert I should get back 0 |
| 21:06:03 | Watusimoto | or maybe a small number if there were a problem at the beginning |
| 21:06:24 | raptor | forgot group by |
| 21:06:40 | raptor | group by player_name |
| 21:06:40 | Watusimoto | did I? |
| 21:06:55 | Watusimoto | I think sam686 should always be authenticated |
| 21:07:03 | raptor | oh |
| 21:07:05 | raptor | sorry |
| 21:07:26 | raptor | i am working with HTTP return codes at work, and i interpreted 451 as something else... :) |
| 21:07:44 | Watusimoto | ah |
| 21:08:16 | Watusimoto | raptor has 56 unauthenticated |
| 21:08:41 | raptor | maybe our authenticated logic is borken? |
| 21:08:45 | raptor | in the game? |
| 21:09:20 | Watusimoto | almost half my entries are unauthenticated |
| 21:10:07 | raptor | ha! |
| 21:10:20 | raptor | all of mine occur from about when we released 016 |
| 21:10:53 | raptor | that woudl sure skew the stats |
| 21:11:08 | raptor | does a game server correctly hold the is_authenticated value? |
| 21:11:11 | Watusimoto | what do you mean? |
| 21:11:25 | raptor | i've noticed some times that people's underlining doesn't show |
| 21:11:25 | Watusimoto | all are clustered around that one date? |
| 21:11:34 | raptor | all are since jan. 29 |
| 21:11:48 | Watusimoto | how about mine and sam's? |
| 21:12:30 | raptor | yours are all since jan. 29 |
| 21:12:35 | raptor | wait! |
| 21:12:38 | raptor | wrong year... |
| 21:13:33 | raptor | most are from jan to april 2011 for you |
| 21:13:35 | Watusimoto | well, player authentication status is very complex |
| 21:13:56 | Watusimoto | any later than april? |
| 21:14:07 | raptor | SELECT insertion_date |
| 21:14:08 | raptor | FROM `stats_player` |
| 21:14:10 | raptor | WHERE player_name = 'watusimoto' |
| 21:14:11 | raptor | AND is_authenticated =0 |
| 21:14:13 | raptor | ORDER BY insertion_date |
| 21:14:14 | raptor | LIMIT 0 , 300 |
| 21:15:09 | raptor | same with me: most are last year jan - may |
| 21:15:31 | raptor | you have only one instance since release of 016, i have none |
| 21:15:33 | Watusimoto | ok |
| 21:15:47 | Watusimoto | well they are ongoing, so whatever the issue is hasn't been fixed |
| 21:16:02 | Watusimoto | I've not really played since 016 |
| 21:16:13 | sam686 | maybe there was a problem with authentication back then (including not showing underlined names when connecting through punching through firewall) |
| 21:16:23 | raptor | sam has had 6 instances since 016 |
| 21:16:34 | Watusimoto | they seem to be grouped |
| 21:16:49 | raptor | maybe on a particular server |
| 21:16:50 | raptor | ? |
| 21:16:53 | Watusimoto | which suggests that auth has failed on a server for some reason and doesn't get corrected |
| 21:17:02 | Watusimoto | maybe... |
| 21:18:06 | Watusimoto | luckily, we can probably correct the historical stats pretty easily if we want |
| 21:18:22 | Watusimoto | since we have a list of everyone who is authenticated |
| 21:19:14 | Watusimoto | unless a server is hacked, names that are authenticated once should always be authenticated, at least since the date of first authentication |
| 21:20:09 | sam686 | i think there is code in master that prevents faking authentication = true in game reports (as in, master checks if that id and player name is connected to master) |
| 21:20:18 | raptor | i think there is a bug somewhere.. |
| 21:21:17 | Watusimoto | I think so too |
| 21:23:47 | raptor | more analysis: http://pastie.org/3536552 |
| 21:24:11 | raptor | number of 'infringements' per server |
| 21:24:28 | Watusimoto | my my guess is that this only happens on remote servers |
| 21:24:40 | raptor | i don't think so |
| 21:24:47 | Watusimoto | or perhaps strike the word only |
| 21:24:48 | raptor | 136 counts on Sam Test |
| 21:24:59 | Watusimoto | but is sam hosting sam test? |
| 21:25:00 | raptor | that seems like it would be local |
| 21:25:11 | raptor | sam686? did you host Sam Test? |
| 21:25:25 | raptor | also 'barf's home' is always local |
| 21:25:29 | Watusimoto | or run it as a dedicated server on your local machine? |
| 21:25:42 | sam686 | i sometimes host Sam Test (if my ip address is 208.107.12.78) |
| 21:25:53 | sam686 | but i don't always host "sam test" |
| 21:26:49 | raptor | i know for sure that 'stay away i'm editing' was hosted through the editor by me :) |
| 21:26:52 | Watusimoto | but clearly many of those are not servers managed by one of us; i.e. they are at least partially remote servers |
| 21:26:59 | raptor | yes |
| 21:27:13 | Watusimoto | games done from editor should not have stats logged for them, right? |
| 21:27:20 | raptor | not since 016, no |
| 21:27:25 | Watusimoto | ah |
| 21:27:32 | Watusimoto | before yes? |
| 21:27:35 | raptor | correct |
| 21:27:49 | raptor | and it wasn't corrected until we almost released 016 |
| 21:31:02 | Watusimoto | it's probably not a problem with remoteclientinfos |
| 21:36:10 | Watusimoto | but I'm kind of at a loss right now on how to figure out where the problem is |
| 21:36:16 | raptor | umm |
| 21:36:30 | raptor | i think some are showing up as authenticated as well that shouldn't be |
| 21:36:41 | raptor | run: |
| 21:36:43 | raptor | select distinct player_name |
| 21:36:45 | raptor | from stats_player |
| 21:36:46 | raptor | where is_authenticated = 1 |
| 21:36:54 | raptor | and you'll see some name that shouldn't be there.. |
| 21:38:58 | Watusimoto | like anyone .0 |
| 21:39:25 | raptor | like 'ihateyouyamenu13' |
| 21:39:44 | raptor | .0 is probably another bug |
| 21:40:04 | raptor | because if you r client crashes, you come back and the server renames you as .0 even though you are authenticated |
| 21:40:14 | raptor | but then it sends the re-name to master at the end |
| 21:46:20 | Watusimoto | but these can't be mostly crash bugs |
| 21:47:07 | raptor | i wonder if there is a problem with pack/unpack update |
| 21:47:13 | raptor | something not entirely in sync |
| 21:47:39 | Watusimoto | there are checks |
| 21:47:54 | Watusimoto | if they are not in sync bit-for-bit, it asserts |
| 21:48:16 | raptor | oh yeah... but what does it do when compiled without TNL_DEBUG? |
| 21:49:30 | Watusimoto | good question |
| 21:49:36 | raptor | try the best it can? |
| 21:50:10 | raptor | maybe we should have sam686 or someone put up a 'Crash Me' server |
| 21:50:16 | raptor | that runs in gdb |
| 21:50:54 | Watusimoto | I don't think it's a bit problem |
| 21:51:01 | Watusimoto | all the other stats look uncorrupted |
| 21:51:40 | Watusimoto | there are enough places where weird stuff can creep in with the authentication |
| 21:51:59 | raptor | creep creep |
| 21:52:34 | | BFLogBot - Commit 220e68ca2dc7 | Author: watusim...@bitfighter.org | Log: Add special case assert to try to trap auth bug |
| 21:53:06 | Watusimoto | at least with this checkin we'll see the problem if it happens on our servers |
| 21:53:29 | Watusimoto | not sure if that will really help or not |
| 21:53:51 | Watusimoto | I'm falling alseep; that;s the best I can manage right now |
| 21:53:55 | raptor | ok |
| 21:53:59 | raptor | good night |
| 21:54:04 | sam686 | ok.. |
| 21:54:47 | raptor | i'll still be here for a while.. |
| 23:48:54 | | Watusimoto Quit (Ping timeout: 272 seconds) |
