Crash report for linux servers

[Phat Looser]

How can I send crash reports if the server is running on linux?

[x86]

You are able to use GDB, but you have to start it before you start the server.

Example: $ gdb mta-server

Type "r" and hit the enter button.

If the server crashes you have to type: "bt"

Pastebin the output.

[Phat Looser]

Not really a pastebin, but:

#0  0xf7183065 in ProcessOfflineNetworkPacket () 
   from /emul/ia32-linux/mta-servers/DgtPuma/net.so 
#1  0xf71840ff in ProcessNetworkPacket () 
   from /emul/ia32-linux/mta-servers/DgtPuma/net.so 
#2  0xf7184e5e in RakPeer::RunUpdateCycle () 
   from /emul/ia32-linux/mta-servers/DgtPuma/net.so 
#3  0xf7187d11 in UpdateNetworkLoop () 
   from /emul/ia32-linux/mta-servers/DgtPuma/net.so 
#4  0xf74d2195 in start_thread () from /lib32/libpthread.so.0 
#5  0xf75bffee in clone () from /lib32/libc.so.6 
  

I know this error from somewhere, but it appears to be on one server only...

This one just happened, it seems to be the same

  
#0  0xf7146065 in ProcessOfflineNetworkPacket ()ady. 
   from /emul/ia32-linux/mta-servers/DgtPuma/net.so 
#1  0xf71470ff in ProcessNetworkPacket () 
   from /emul/ia32-linux/mta-servers/DgtPuma/net.souma/mta-server 
#2  0xf7147e5e in RakPeer::RunUpdateCycle () 
   from /emul/ia32-linux/mta-servers/DgtPuma/net.so 
#3  0xf714ad11 in UpdateNetworkLoop () 
   from /emul/ia32-linux/mta-servers/DgtPuma/net.so 
#4  0xf7495195 in start_thread () from /lib32/libpthread.so.0 
#5  0xf7582fee in clone () from /lib32/libc.so.6 
  

[x86]

Are you using zombie mode ?

[Phat Looser]

That one isn't a server I administrate - I just host it, it belongs to DgtPuma.

My own servers never had that issue.

His default gamemode is race, and every time I connected to it there was no other gamemode running than race.

The administrator of the server is bandi93.

As far as I know his race gamemode is modded, but since he never asked for it I never looked through his code.

Weird thing is that his server stopped crashing after I cleaned up his databases.

A backup of the database doesn't work and results in a crash too. But I really don't know if the one has to do with the other.

[ccw]

Do you apply memory limits?

[Phat Looser]

Nope, I don't. I think 12 Gig of RAM is enough for 10 servers

Last time I checked his system he was using less then 10%, that means less then 1 Gigabyte of RAM.

His CPU usage was below the 20% limit (for one core), too.

Those limits are pure fictonal and just result in working together until the LUA scripts are optimized. There is no software limit. If someone goes above those limits, I'll just be nagging them to death.

So, his resource usage was "quite normal" for a race server with 20 people.

[Phat Looser]

Another one is reporting crashes:

#0  0xf7480d7f in raise () from /lib32/libc.so.6 
#1  0xf7482788 in abort () from /lib32/libc.so.6 
#2  0xf74bd6f5 in ?? () from /lib32/libc.so.6 
#3  0x0000002e in ?? () 
#4  0xf2443e84 in ?? () 
#5  0x00000400 in ?? () 
#6  0xf0900000 in ?? () 
#7  0xf75a3ff4 in ?? () from /lib32/libc.so.6 
#8  0xf758acf8 in ?? () from /lib32/libc.so.6 
#9  0x00000017 in ?? () 
#10 0xff84fb20 in ?? () 
#11 0x00000031 in ?? () 
#12 0xf758ad11 in ?? () from /lib32/libc.so.6 
#13 0x00000002 in ?? () 
#14 0xf758add0 in ?? () from /lib32/libc.so.6 
#15 0x0000001f in ?? () 
#16 0xf758ad15 in ?? () from /lib32/libc.so.6 
#17 0x00000004 in ?? () 
#18 0xf24443b3 in ?? () 
#19 0x00000008 in ?? () 
#20 0xf758ad1b in ?? () from /lib32/libc.so.6 
#21 0x00000005 in ?? () 
#22 0x0ac023c0 in ?? () 

I dunno, but it COULD be someone is shutting down his server, so I don't know if its an MTA bug.

Only thing I know that his logfile is 1.8 GB of scripting errors.

€dit:

It happens over and over, even though I blocked the contact to that server.

It seems the mass of "illegal object" errors it throws crashes the server over and over, always with the same error (above).

I'll force the scripts deactivated and I'll restart the server.

#0  0xf746bd7f in raise () from /lib32/libc.so.6 
#1  0xf746d788 in abort () from /lib32/libc.so.6 
#2  0xf74a86f5 in ?? () from /lib32/libc.so.6 
#3  0x00000030 in ?? () 
#4  0xf1efde84 in ?? () 
#5  0x00000400 in ?? () 
#6  0x00000031 in ?? () 
#7  0x0a5e8ab0 in ?? () 
#8  0xf7575cf8 in ?? () from /lib32/libc.so.6 
#9  0x00000017 in ?? () 
#10 0xfffcab20 in ?? () 
#11 0x00000031 in ?? () 
#12 0xf7575d11 in ?? () from /lib32/libc.so.6 
#13 0x00000002 in ?? () 
#14 0xf7575dd0 in ?? () from /lib32/libc.so.6 
#15 0x0000001f in ?? () 
#16 0xf7575d15 in ?? () from /lib32/libc.so.6 
#17 0x00000004 in ?? () 
#18 0xf1efe3b3 in ?? () 
#19 0x00000008 in ?? () 
#20 0xf7575d1b in ?? () from /lib32/libc.so.6 
#21 0x00000005 in ?? () 
#22 0x0a6accf0 in ?? () 
  

Its a "sigabort" that was sent.

If its a bug, I'll make it a mantis report.

€dit2:

Its a bug. Not a feature.

[Phat Looser]

Trying to figure out this one:

on start:

http://pastebin.com/6rcSEe6A

on map change (depends on the map thats loaded)

http://pastebin.com/yADp7bZe