Project Reality Forums

Posted: **2011-03-05 10:59**

Soppa wrote: [..]

Server is also experiencing this mapload issue which could cause some of problems.
I dont know anything about it and somehow I cant find Server Administrator area where could be solution for it. If I try to push apply for server license, it just says that I already have license.

There may have gone something wrong. You are not listed as serveradmin either.
I think this mapchange thing is a bug on 64bit Linux, there are a few admins with this issues two and I dont think somebody found a fix for that. But it also doesnt look like this is causing more issues.

I sent you a PM with 2 links, feel free to contact me with questions serverwise.

echo wrote:When the official test was done on the NwA server,[...]

Wasnt it on NEW server

?

Posted: **2011-03-05 10:59**

You should be able to see the SA forums now.

Posted: **2011-03-05 12:30**

BloodyDeed wrote: I think this mapchange thing is a bug on 64bit Linux, there are a few admins with this issues two and I dont think somebody found a fix for that. But it also doesnt look like this is causing more issues.

This is what happenes just before that "delay".. mprotect is called

Code: Select all

[pid 31815]      0.000035 times({tms_utime=32523, tms_stime=231, tms_cutime=0, tms_cstime=0}) = 1607539251
[pid 31815]      0.000068 mprotect(0x2aaac5ac3000, 114688, PROT_READ|PROT_WRITE) = 0
[pid 31815]      0.008068 mprotect(0x2aaac5adf000, 524288, PROT_READ|PROT_WRITE) = 0
[pid 31818]      0.007957 <... select resumed> ) = 0 (Timeout)
[pid 31818]      0.000039 select(22, [21], NULL, NULL, {3, 0} <unfinished ...>
[pid 31818]      0.004946 <... select resumed> ) = 0 (Timeout)

<<< BREAK where CPU load goes high and server sits doing nothing for 1min
after this weird wait time it will continue load map without problem and does it pretty fast >>>

[pid 31818]      0.000041 select(22, [21], NULL, NULL, {3, 0} <unfinished ...>
[pid 31818]      0.006002 <... select resumed> ) = 0 (Timeout)
[pid 31818]      0.000045 select(22, [21], NULL, NULL, {3, 0} <unfinished ...>
[pid 31818]      0.002997 <... select resumed> ) = 0 (Timeout)
[pid 31818]      0.000044 select(22, [21], NULL, NULL, {3, 0} <unfinished ...>
[pid 31815]      0.006507 munmap(0x2aaab1756000, 2105344) = 0
[pid 31815]      0.001656 close(23)     = 0
[pid 31815]      0.000043 munmap(0x2aaaaaab1000, 4096) = 0
[pid 31815]      0.000712 open("mods/pr/gamelogicinit.con", O_RDONLY) = 23
[pid 31815]      0.000064 lseek(23, 0, SEEK_CUR) = 0
[pid 31815]      0.000030 lseek(23, 0, SEEK_END) = 323
[pid 31815]      0.000028 lseek(23, 0, SEEK_SET) = 0
[pid 31815]      0.000128 read(23, "rem *** Put non level specific i"..., 323) = 323

and also cpu timings of bf2 server including mapchange:

Code: Select all


% time     seconds  usecs/call     calls    errors syscall
------ ----------- ----------- --------- --------- ----------------
 57.84    0.001657           0     32991           read
 33.58    0.000962           0     42967           lseek
  4.47    0.000128           0       601           brk
  1.64    0.000047           0       120           sendto
  1.26    0.000036           0      1109      1102 open
  0.49    0.000014           1        17        13 stat
  0.38    0.000011           0       262           rt_sigaction
  0.35    0.000010           0      1319      1210 recvfrom
  0.00    0.000000           0        52           write
  0.00    0.000000           0         7           close
  0.00    0.000000           0        10           fstat
  0.00    0.000000           0         6           mmap
  0.00    0.000000           0         1           mprotect
  0.00    0.000000           0         5           munmap
  0.00    0.000000           0         1           ioctl
  0.00    0.000000           0      3080           select
  0.00    0.000000           0      1420           nanosleep
  0.00    0.000000           0         2           socket
  0.00    0.000000           0      1212      1212 accept
  0.00    0.000000           0         2           bind
  0.00    0.000000           0         2           setsockopt
  0.00    0.000000           0         1           clone
  0.00    0.000000           0         4           fcntl
  0.00    0.000000           0       397           times
------ ----------- ----------- --------- --------- ----------------
100.00    0.002865                 85588      3537 total

Posted: **2011-03-05 18:24**

I saw the server up just now and it disappeared, crash??

Posted: **2011-03-05 18:32**

crot wrote:I saw the server up just now and it disappeared, crash??

no, its that damn mapchange which takes 2-3min

I made some research about this, it seems to be a bug in kernel on memory managing.
They have found solution half year ago but have no clue which kernel has patched.

Could some admin try with newest kernel on 64bit linux how this works..
Maybe I should post this at SA area

Posted: **2011-03-05 20:32**

Thanks for joining today. We did catch another problem which is causing crash on mapchange.
I will give a look for it next.

Module: Common
File: Game/Common/GhostManager.cpp
Line: 245
Text: Out of ghost ids, max:1023

Sounds funny error

Posted: **2011-03-05 20:56**

Soppa wrote:Thanks for joining today. We did catch another problem which is causing crash on mapchange.
I will give a look for it next.

Module: Common
File: Game/Common/GhostManager.cpp
Line: 245
Text: Out of ghost ids, max:1023

Sounds funny error

Ive gotten that error on my server aswell, im guessing its common PR crash. Nothing to worry about.

Sounds like it has to many playable objects or smt though.

EDIT: Just googled it, and whataya know.

https://www.realitymod.com/forum/f27-pr ... erver.html

BTW, im running a Windows 64 bit server, so its a general issue im guessing.

Qouting a thread here
"...spearheading a movement to get the game renamed to Battlefield CTD. Battlefield CTD (or Battlefield Crash To Desktop for those who enjoy the extra reading), is a incredibly fun game when it works. Unfortunately, this title has that rushed to deadline feel to it and it shows. The most common crash is the now infamous GhostManager.cpp error. For those without programming experience the *.cpp extension is industry standard for C Plus Plus, the programming language the game was coded in. The Ghost Manager is the part of the game that handles the respawning of dead players."

So respawning of dead players. mmmm Yeah makes sence?

Posted: **2011-03-05 21:26**

Wicca wrote:Ive gotten that error on my server aswell, im guessing its common PR crash. Nothing to worry about.

ok thanks, lets try to repair it as well.

Posted: **2011-03-05 22:06**

Soppa wrote:Thanks for joining today. We did catch another problem which is causing crash on mapchange.
I will give a look for it next.

Module: Common
File: Game/Common/GhostManager.cpp
Line: 245
Text: Out of ghost ids, max:1023

Sounds funny error

Maybe break this limit

??
(Each dynamically created objects = ghosts)

Posted: **2011-03-05 22:09**

Server software or w/e that is, is super old.. cant even get on a tank .50cal (black screen), can u update to like 0.95e?

Posted: **2011-03-05 22:51**

Seems it crash after two mapchange for that ghost thingy.
Will try to fix it tomorrow. Ill just start for Qwai for now if peps still are willing to go for it.

Update: Seems Qwai segfaulted server. Thanks who ever were playing there and if you got any information what you did when this happened, please reply to this.

Got dump and will check it out.

Ill take server now offline until sorted these few things. Thanks for everyone who joined tonight and yesterday.

Posted: **2011-03-05 22:52**

Okay there needs to be some kind of teamwork rule or incentive or something. I just jumped in, and no matter what squad I join, there was nothing but a clusterfuck. Squad leaders sleeping or not communicating at all and just leaving without notice, no coordination. It was like a 126 man pubberfest.

Posted: **2011-03-06 00:42**

Get an admin team, make rules = profit?

Signups where?

Posted: **2011-03-06 01:16**

Will the FH2 server be up tomorrow again? were there any technical problems with it except squad bug?

Posted: **2011-03-06 02:55**

He said that it would be up and that he would get Purple Heart Lane to run

Thanks Soppa for the FH2 server. Awesome job, no crash, only squad bug

Posted: **2011-03-06 03:03**

Pvt.LHeureux wrote:He said that it would be up and that he would get Purple Heart Lane to run

Thanks Soppa for the FH2 server. Awesome job, no crash, only squad bug

Well this is not right place for FH2 talks

??:
But yes, it will be up on sunday but not from monday so have fun

Posted: **2011-03-06 03:05**

Tema567 wrote:Maybe break this limit ??
(Each dynamically created objects = ghosts)

I made first fix try for this "Out of ghost ids" bug.

Server is up and running so peps go crash it

Posted: **2011-03-06 05:14**

Steakslim wrote:Okay there needs to be some kind of teamwork rule or incentive or something. I just jumped in, and no matter what squad I join, there was nothing but a clusterfuck. Squad leaders sleeping or not communicating at all and just leaving without notice, no coordination. It was like a 126 man pubberfest.

That's because it's FH2 not PR.

Posted: **2011-03-06 05:30**

much fun on 128 ?

Posted: **2011-03-06 11:19**

ok heres what I think;

1) this only works on big maps, otherwise too much server warping
2) matches are much more epic, there is real suppression going on, by this mean, there is a more pronounced front line, where the people are litterally suppressed back, i think this adds a more realistic feel
3) amount of armour is good as it is, just need more transport
4) hardly any FPS drop for me, just mainly server warping lag on smaller maps
5) feels so much better, can deploy more troops to defend and not just the typical one squad with six people..
6) communication an issue, but i think taht will resolve it self once people adapt and use things like mumblemore.

7) pure epicness and real step forward.

Project Reality Forums

128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations

Re: 128 player tests: Feedback and Observations