=[JFF]=UDieToday-Vince

Members
  • Content Count

    23
  • Joined

  • Last visited

  • Days Won

    1

=[JFF]=UDieToday-Vince last won the day on January 23

=[JFF]=UDieToday-Vince had the most liked content!

Community Reputation

2 Neutral

About =[JFF]=UDieToday-Vince

  • Rank
    Member

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Not an expert on the lua implementation but I think the readme.md has a problem I found this, https://github.com/leiizko/cod4x_lua_plugin/search?q=mysqrt&type=Code to me, it's likely something funny in the repository because this code looks like it would work: https://github.com/leiizko/cod4x_lua_plugin/blob/2f63101c82a752d6831d45744949b421e20421c4/README.md Plugin_ScrAddFunction( "mysqrt" ) 31 32 function mysqrt () 33 local n = Plugin_Scr_GetFloat( 0 ) --get number (float) from gsc … 53 num = 123; 54 print( "Square root of " + num + " is " + mysqrt( num ) + "\n" ); 55 }
  2. That looks like a fine idea also, but I am thinking that the activity or writing the script error and few other lines of what called it might have a performance impact on one of these loops, somewhat proven by turning off the writing with good results. My preference would be to add a log level flag, that would allow a developer to enhance the logging, but otherwise provide highest level of game performance.
  3. I took it a step further: scr_vm_main.c line 1055 I commented the whole block as shown below, we had 15 players with ~2500 zombies in the finale and have not had a suspected infinite loop in a week. thanks for the point in the right direction guys, much appreciated for all the work and help Vince void RuntimeError_Debug(char *msg, char *pos, int a4) { /* int i; Com_Printf(CON_CHANNEL_SCRIPT,"\n^1******* script runtime error *******\n%s: ", msg); Scr_PrintPrevCodePos(0, pos, a4); if (gScrVmPub.function_count) { for (i = gScrVmPub.function_count - 1; i > 0; --i) { Com_Printf(CON_CHANNEL_SCRIPT,"^1called from:\n"); Scr_PrintPrevCodePos(0, gScrVmPub.function_frame_start.fs.pos, gScrVmPub.function_frame_start.fs.localId == 0); } Com_Printf(CON_CHANNEL_SCRIPT,"^1started from:\n"); Scr_PrintPrevCodePos(0, gScrVmPub.function_frame_start[0].fs.pos, 1); } Com_Printf(CON_CHANNEL_SCRIPT,"^1************************************\n"); */ }
  4. You are trying to compile a custom version ? Or just get a server up fast and easy on linux ? I found linuxgsm an easy way to get my servers up and running was linux gsm which is free and helps with setting up cod4x... There are pretty good instructions here if you are stuck... https://linuxgsm.com/lgsm/cod4server/
  5. Agreed, however I have fixed about 15 or 20 spots in the scripts and they just keep coming... So my thought was to use a bigger hammer and perhaps disable the logging and see if the loop finished in time. I do not think it is actually endless, just think that the time to write all those errors might make it appear so.
  6. Pretty much what happens is that the mod starts to throw some errors like below: each error has quite a bit of information. I have worked on checking for undefined and other issues but the mod is somewhat riddled and some still remain, I have had some luck but think that the writing of all this data is partly responsible for the time elapsed. So my understanding is that https://github.com/callofduty4x/CoD4x_Server/blob/b9a296ab9c3ffcec34cb1d560c5b4475c57fd3bc/src/scr_vm_main.c#L1372 there are checks for a thread running longer than 6s, I have modified the code to go for 10s but did not really solve the problem. My idea is that the logging of all this data below is quite possibly a major contributor to the length of time and that if I could add a flag to turn off this level of logging it might just work. Here is a sample error, looped in an area for a while hit the time limit and did a map rotate. The ROTU guys are working on 1.0 but not certain that will come soon. [1;31m******* script runtime error *******[0mcannot cast undefined to string: ] (file 'scripts/players/_usables.gsc', line 190)[0m ] self.hinttext setText(ent.hintstring);[0m ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] *[0m ] [1;31mcalled from:[0m ] (file 'scripts/players/_usables.gsc', line 124)[0m ] if (getBetterUseObj(dis))[0m ] ] ] ] ] ] ] ] ] *[0m ] [1;31mstarted from:[0m ] (file 'scripts/players/_usables.gsc', line 155)[0m ] wait .05;[0m ] ] ] ] *[0m ] [1;31m************************************[0m ] [1;31mError: script runtime error: potential infinite loop in script - killing thread.[0m ] (file 'scripts/players/_usables.gsc', line 83)[0m ] while (1)[0m ] ] *[0m ] [1;36mHitch warning: 10036 msec frame time[0m ] map_rotate...[0m
  7. We are running a rotu server, sometimes we have some issues when the code runs over the next frame... I have been thinking about it a bit and think that the time to write errors to the log might be a large part of what is causing the behavior. The mod is here: https://rotu-revolution.com/ The mod has bugs, and my attempts to fix seem to bring some minor help but not a permanent solution. My question: Is there a way to disable or lower the logging level for the errors ? I have a feeling that the problems with the mod would not overflow the frame if was not disk write to each line. Little hacky, but want something till 1.0 of the mod finally comes out.
  8. I am observing this with some frequency in my logs as well: bad connectionless packet from 188.165.57.239:27953 When I review the logs, I can find that ip and also that port: Master0: cod4master.cod4x.me Resolving cod4master.cod4x.me cod4master.cod4x.me resolved to 188.165.57.239:20810 and this: Resolving cod4update.cod4x.me cod4update.cod4x.me resolved to 188.165.57.239:27953 Any idea if anything clientside is problem ? thanks
  9. I've actually seen the same behaviour, also running new experience. Vince
  10. Rotu Revolution, Click To Join -- 18.185.183.228:28960 New Experience - Click To Join -- 3.123.242.205:28960 Softcore - Click To Join -- 18.196.25.40:28962 Promod - Click To Join -- 18.185.183.228:28961 Link to Search GameTracker for =[JFF]= All are welcome, we are an active community, been around > 10 years and are accepting new members and people looking to help out. For more details visit our forum here Hope to see you in game Vince
  11. Not quite sure how this came about or how to solve it.. but I now have 2 serverfiles directories /serverfiles which was populated with data up until about a month ago /serverfiles  -- which began populating after then... the i and >> and upside ? are all part of the folder name... I have examined the start commands, etc. and do not find anything which explains, can anyone provide some insight on how to troubleshoot ? Game is running fine, just funny to have this duplicate folder Vince
  12. We were asking ourselves the same question, I found this https://github.com/byNeHo/CoD4x-WebAdmin and we like it a lot.
  13. Old post, but I had success executing the sql here: https://github.com/BigBrotherBot/big-brother-bot/blob/master/b3/sql/mysql/b3-update-1.8.1.sql Not sure xlrstats is a good solution going forward, from what I found it requires a php version unsupported since 2010... And unless you are sharp on updating the version of php cake that it was written with, might take some work... Good luck, and hope to see you in game...
  14. Not sure if this code is going forward with the botnames.txt, I am using sometimes with [BOT] to not fool gametracker, I found that the bot names were being repeated sometimes... I believe that the below line with the rand() % cntnames with a small number of bot names will result in duplicates... In this case, would it be preferable to have same line with variable i ( bot # ) as an option in the code if going forward ? And from the poster above he is using 40 out of 240, but think just had 128 so highly likely he would have duplicates with that number of bots. Q_strncpyz(name,botnames[rand() % cntnames],sizeof(name)); Much appreciated for all the efforts and help that all have given me Vince
  15. @volkv and all, I recently had an issue where a player joined and when we tried getss immediately left, about 10x... I am planning to implement the code you shared, but don't see anything around this logging that you mentioned.... Are you writing this somewhere this is not in the code ? thanks Vince