Server Crash stats?

[Tommygun]

Would it be worthwhile having a record of server crashes, time, # of players server is set to and stuff?
Anecdotal evidence suggests (to me) that the New Era (German) server seems to be less crash prone than others (as one might expect seeing the way the country is run and the trains come on time). I would expect an equal # crashes between servers with the same settings but to me it seems nwa and merk fall over a bit more often (It may be I just play on them more often). (oh, and no slight intended, I love both and appreciate all the work)
Doing this we may spot a trend? one could report here and just use this webpage time zone info for consistency. Maybe its a **** idea or others have better ideas.

[Mineral]

Private server admin forums have a thread like that with SA giving crashlogs and so on. But yeah, this is the things we need cheers tommy

[ghostfool84]

As regular player on [NEW] i think it crashes like one time every evening. I dont think other servers like Merk crash more often.

Some kind of an automatic crash agent that sends in a error report with infos that DEVs can use for troubleshooting would help to record all the crashes, that could be really useful i think. I dont thik every crash that could be useful is listed in that crashlog thread.

[Solid Knight]

CIA crashes all the time. Over the last two weeks it's averaged three map changes before a crash and sometimes multiple times in a row on the first map change. HOG crashes a lot less, averaging about six map changes before a crash.

I doubt it's the servers being low quality but rather something in the PR server that causes it just like how certain things trigger CTD on the client side.

[SyntheticCoyote]

CIA has been crashing. We are working on it gents. There is a thread that the comm leaders use for crashes with the Devs. We are on it.

[Prevtzer]

What's the reason for keeping these logs and info private? Are you afraid somebody might help with the crashes? Same with the bug logs...

[Mineral]

As it's where they belong? Each thing has it's place. Nobody was forced to put them there. It's just where they logically belong.

[Prevtzer]

As it's where they belong? Each thing has it's place. Nobody was forced to put them there. It's just where they logically belong.

They belong to a private part of a forum? If we go by that logic they belong to the Technical support but that's besides my point anyway. I don't see any reason to keep them private as somebody from the community might just help with the issues. You're obviously not doing such a good job since these threads pop up every few days, so you could really use some help.

[FatalWaffles]

Prevtzer: It really would only make sense for them to be placed in the server administrators forum. It is an issue that the server administrators are dealing with and it is an uncluttered and easier to view place to allow server administrators to easily communicate amongst themselves and communicate with the development team about issues with the servers.

I am confident that the development team is working to solve the issues and is probably working with their team of testers to help troubleshoot and solve the issues as well. I know there is some communication between the development team and the server administrators so it shows at least some progress is being made.

If you, or anyone else for that matter, would like to help out I'm sure they wouldn't mind. I believe you can still apply to be on the PR:BF2 Testing Team and I'm sure that if you have something valuable to offer to the development team you can reach out and offer your assistance to them. I'm sure that most, if not all, of the developers and contributors do this in their spare time and not as a full time job. After all it is a free mod.

[Prevtzer]

Prevtzer: It really would only make sense for them to be placed in the server administrators forum. It is an issue that the server administrators are dealing with and it is an uncluttered and easier to view place to allow server administrators to easily communicate amongst themselves and communicate with the development team about issues with the servers.

Well, whoever is dealing with it isn't making much of a progress, I don't see why these logs shouldn't be publicly published so maybe, just maybe the community could help as well. You can keep your empty server admin thread, but make a public one as well.

If you, or anyone else for that matter, would like to help out I'm sure they wouldn't mind. I believe you can still apply to be on the PR:BF2 Testing Team and I'm sure that if you have something valuable to offer to the development team you can reach out and offer your assistance to them. I'm sure that most, if not all, of the developers and contributors do this in their spare time and not as a full time job. After all it is a free mod.

You can't apply, I tried.

Just make the damn logs public, it's good for everyone.

[brezmans]

A crash log was posted in a very similar thread to this one:

https://www.realitymod.com/forum/f10-pr ... ost1949781

Start cracking:

Code: Select all

2013-09-14-20-54-25
[New LWP 27918]
[New LWP 27920]
[New LWP 27919]
[New LWP 27914]
Core was generated by `/opt/bf2skirmish1/bin/amd-64/prbf2_l64ded +modPath mods/pr'.
Program terminated with signal 11, Segmentation fault.
#0 0x00000000004d3e27 in ?? ()
core
[New LWP 27918]
[New LWP 27920]
[New LWP 27919]
[New LWP 27914]
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
Core was generated by `/opt/bf2skirmish1/bin/amd-64/prbf2_l64ded +modPath mods/pr'.
Program terminated with signal 11, Segmentation fault.
#0 0x00000000004d3e27 in dice::hfe::CheckServerAliveThread::run() ()
#0 0x00000000004d3e27 in dice::hfe::CheckServerAliveThread::run() ()
No symbol table info available.
#1 0x000000000074afaa in dice::hfe:anonymous namespace):threads_thread_trampoline(void*) ()
No symbol table info available.
#2 0x00007f312b3a6fb6 in start_thread () from /lib64/libpthread.so.0
No symbol table info available.
#3 0x00007f312a8ba48d in clone () from /lib64/libc.so.6
No symbol table info available.
===================

[SyntheticCoyote]

@ Brez

[Tommygun]

Im kinda with Prevtzer...these Eastern Europeans seem to have flair when it comes to geeking. Sure keep them separate forums so you admins don't get a cluttered thread but bounce them to an open forum for those who have spme interest after all its why Linux rocks coz there are so many people helping and developing. Maybe no one will help but there may be one extra bit of insight from someone and whats the harm, we all want this thing to be the best it can be?

[FatalWaffles]

Im kinda with Prevtzer...these Eastern Europeans seem to have flair when it comes to geeking. Sure keep them separate forums so you admins don't get a cluttered thread but bounce them to an open forum for those who have spme interest after all its why Linux rocks coz there are so many people helping and developing. Maybe no one will help but there may be one extra bit of insight from someone and whats the harm, we all want this thing to be the best it can be?

Core dumps from crashes have been posted above and elsewhere in public view. Little to no help/assistance has been given...

[Prevtzer]

Core dumps from crashes have been posted above and elsewhere in public view. Little to no help/assistance has been given...

Maybe try hiding them a bit more. Page 109 of a non-relevant thread would be a great place to put them. 1.0 has been out for some months now and all of the major problems still haven't been successfully dealt with, additional help sure seems like a good option.

Im kinda with Prevtzer...these Eastern Europeans seem to have flair when it comes to geeking. Sure keep them separate forums so you admins don't get a cluttered thread but bounce them to an open forum for those who have spme interest after all its why Linux rocks coz there are so many people helping and developing. Maybe no one will help but there may be one extra bit of insight from someone and whats the harm, we all want this thing to be the best it can be?

Who are you calling an Eastern European?

[FatalWaffles]

Maybe try hiding them a bit more. Page 109 of a non-relevant thread would be a great place to put them. 1.0 has been out for some months now and all of the major problems still haven't been successfully dealt with, additional help sure seems like a good option.

There is a core dump a few posts above and that post even links to another one. Get cracking.

[brezmans]

Maybe try hiding them a bit more. Page 109 of a non-relevant thread would be a great place to put them. 1.0 has been out for some months now and all of the major problems still haven't been successfully dealt with, additional help sure seems like a good option.

The logs are here now, I posted one, what else do you want?

Here's a second one, just to make sure you are fully satisfied with the availability of the server crash logs. So better start solving it, 1.0 has been out for some months now.

Code: Select all

Program terminated with signal 11, Segmentation fault.
#0  0x00000000033b3390 in ?? ()
(gdb) bt
#0  0x00000000033b3390 in ?? ()
#1  0x00007fec8886ac73 in eval_frame ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#2  0x00007fec8886b5f5 in eval_frame ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#3  0x00007fec8886cf30 in PyEval_EvalCodeEx ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#4  0x00007fec8886cff4 in PyEval_EvalCode ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#5  0x00007fec8888510d in run_node ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#6  0x00007fec88864dac in builtin_eval ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#7  0x00007fec8886b4df in eval_frame ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#8  0x00007fec8886b5f5 in eval_frame ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#9  0x00007fec8886cf30 in PyEval_EvalCodeEx ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#10 0x00007fec88898f66 in function_call ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#11 0x00007fec8888899e in PyObject_Call ()
   from /home/pr/public/bin/amd-64/libdice_py.so
---Type <return> to continue, or q <return> to quit---
#12 0x00007fec88867dfb in PyEval_CallObjectWithKeywords ()
   from /home/pr/public/bin/amd-64/libdice_py.so
#13 0x00000000004febcf in dice::hfe: :p ython: :P ythonHost::handleEvent(dice::hfe::ScriptEvent, _object*) ()
#14 0x00000000004a664e in dice::hfe::ServerGameLogic::handlePickup(dice::hfe::world::IObject*, dice::hfe::world::IPlayer*, dice::hfe::world::IObject*, bool) ()
#15 0x00000000004aa8b1 in dice::hfe::ServerGameLogic::spawnPlayer(dice::hfe::world::IPlayer*, std: :p air<dice::hfe::Vec3, dice::hfe::Vec3>*) ()
#16 0x00000000004ab736 in dice::hfe::ServerGameLogic::uPlayingSpawning(float)
    ()
#17 0x00000000004ab9e1 in dice::hfe::ServerGameLogic::uPlaying(float) ()
#18 0x00000000004ae5c1 in dice::hfe::ServerGameLogic::update(float) ()
#19 0x000000000045af92 in dice::hfe::GameServer::simulateFrame(float) ()
#20 0x0000000000461324 in dice::hfe::GameServer::update(int, float) ()
#21 0x00000000004db123 in dice::hfe::BF2Engine::mainLoop() ()
#22 0x000000000040b29c in dice::hfe::BF2::run() ()
#23 0x000000000040b868 in main ()

Please, go crazy.

[Prevtzer]

I guess it's pointless to argue with kitties


keep it friendly
~k4on

[brezmans]

I guess it's pointless to argue with kitties


keep it friendly
~k4on

Who's arguing? There are 2 completely different crash logs here. Either point the devs to a solution or accept that these logs are kept private for a reason.

[DDS]

A crash log was posted.. Start cracking:

Who's arguing? There are 2 completely different crash logs here. Either point the devs to a solution or accept that these logs are kept private for a reason.

Core dumps from crashes have been posted above and elsewhere in public view. Little to no help/assistance has been given...

Well to be fair there is little to no context (Date, time, servers specs, #players ect.) given with those logs.

If you are sincere then there should be no problem providing it here for those who want to help.
Unless and until a developer or moderator says otherwise I don't think its in anyone's prerogative to shut down discussion yet again on viewing server crash logs.