Segmentation fault when starting Linux server

[thallanor]

I realize that there are several posts regarding this issue, but none of the proposed solutions seem to work or are applicable in my case. If someone has any ideas, I'd greatly appreciate them. My Linux server uses SME Server, based on CentOS. The server hardware includes dual Xeon 2.0 GHz processors w/ Hyperthreading and 512 MB of RAMBUS RAM and 6 x Seagate Cheetah 36 GB 15K RPM hard drives in a RAID 5 array. If you require any additional information from me, please let me know and I will respond as soon as possible. I am not an expert in Linux, but I can get around and do most of the basics.

Thanks in advance -- I'd greatly appreciate ANY input, advise, or suggestions regarding this issue.

Please find attached both the contents of debug.log and the results of running it at the command line. Note that I have attempted to install and run this both as my local user and as root -- the results are identical. This output was prepared as root.

Sincerely,
Dennis


[root@berio srcds_1]# ./srcds_run -console -game tf -dir . -debug
Auto detecting CPU
Using SSE2 Optimised binary.
Enabling debug mode
Server will auto-restart if there is a crash.

Console initialized.
./srcds_run: line 344: 6339 Segmentation fault (core dumped) $HL_CMD
Cannot access memory at address 0xb7ff9000
Cannot access memory at address 0xbfffd5cc
debug.cmds:3: Error in sourced command file:
Cannot access memory at address 0xb7ff9000
email debug.log to linux@valvesoftware.com
Mon Nov 12 22:02:59 MST 2007: Server restart in 10 seconds
Mon Nov 12 22:03:01 MST 2007: Server Quit


----------------------------------------------
CRASH: Mon Nov 12 21:56:53 MST 2007
Start Line: ./srcds_i686 -console -game tf -dir . -debug
Using host libthread_db library "/lib/libthread_db.so.1".
Core was generated by `./srcds_i686 -console -game tf -dir . -debug'.
Program terminated with signal 11, Segmentation fault.
#0 0x00000000 in ?? ()
#0 0x00000000 in ?? ()
No symbol table info available.
End of Source crash report
----------------------------------------------

[lucindrea]

try this .. it's a bit more informative

Code:

strace ./srcds_run -console -game tf -dir . -debug 2>debug.out

then less through the debug.out .. it's most likely dieing on a lib.

[thallanor]

lucindrea Wrote:try this .. it's a bit more informative

Code:

strace ./srcds_run -console -game tf -dir . -debug 2>debug.out

then less through the debug.out .. it's most likely dieing on a lib.

Thanks! When I did this, I saw the following:

Auto detecting CPU
Using SSE2 Optimised binary.
Enabling debug mode
Server will auto-restart if there is a crash.

Console initialized.
email debug.log to linux@valvesoftware.com
Wed Nov 14 14:35:22 MST 2007: Server restart in 10 seconds

Console initialized.
email debug.log to linux@valvesoftware.com
Wed Nov 14 14:35:32 MST 2007: Server restart in 10 seconds

I am now going through debug.out to see if anything stands out. I'm not sure if I can make anything out, but I'll try!

Thanks again.

[thallanor]

lucindrea Wrote:try this .. it's a bit more informative

Code:

strace ./srcds_run -console -game tf -dir . -debug 2>debug.out

then less through the debug.out .. it's most likely dieing on a lib.

I think you are onto something! I found this:

open("/usr/share/locale/en_US/LC_MESSAGES/libc.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/share/locale/en/LC_MESSAGES/libc.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
write(2, "./srcds_run: line 344: 21396 Seg"..., 75./srcds_run: line 344: 21396 Segmentation fault (core dumped) $HL_CMD
) = 75

Now I need to find out why I am having problems with libc.

[lucindrea]

it's part of the glibc package .. more , it's part of locales .. i dont know what dist you are useing , but fedora/redhat would use yum and debian would use aptget ... either way , see if you have an update for the glibc package.
http://rpmfind.net/linux/RPM/PLD/dists/ac/ready/i386/glibc-2.3.2-15.i386.html


try

Code:

/lib/libc.so.6

and see what you get .. that should tell you your version .. i can tell you that fedora 7 has the following

Code:

[root@lucindrea ~]# /lib/libc.so.6
GNU C Library stable release version 2.6, by Roland McGrath et al.
Copyright (C) 2007 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.
Compiled by GNU CC version 4.1.2 20070502 (Red Hat 4.1.2-12).
Compiled on a Linux 2.6.9 system on 2007-07-08.
Available extensions:
        The C stubs add-on version 2.1.2.
        crypt add-on version 2.1 by Michael Glad and others
        GNU Libidn by Simon Josefsson
        Native POSIX Threads Library by Ulrich Drepper et al
        BIND-8.2.3-T5B
        RT using linux kernel aio
For bug reporting instructions, please see:
<http://www.gnu.org/software/libc/bugs.html>.

PLEASE BE AWARE THAT CHANGING GLIBC CAN BREAK YOUR ENTIRE SYSTEM IF YOUR NOT CARFULL

this is why i suggest useing whatever update program comes with your system .. glibc is a core system lib package .. so double then triple check everything.

[thallanor]

lucindrea Wrote:it's part of the glibc package .. more , it's part of locales .. i dont know what dist you are useing , but fedora/redhat would use yum and debian would use aptget ... either way , see if you have an update for the glibc package.
http://rpmfind.net/linux/RPM/PLD/dists/ac/ready/i386/glibc-2.3.2-15.i386.html

Weird thing is that according to that, I am running the same version. But ./en/ and ./en_US/ are missing the file libc.mo but ./en_GB has it. (I tried to copy the one from ./en_GB into ./en and ./en_US and it didn't report a file missing, but still crashed with a seg fault.) Apparently, I have the newest version for me distro (SME Server, built on CentOS). So I'm not sure what to do. Bleh!

[lucindrea]

well , not finding a lang file may not be your issue either .. could just be thats where it's dieing ..
you could try running strace with the -f flag or -d flag .. but thats ALOT of output .. man strace should give more info ... it looks to me that it'd dieing around that point , but it could be whatever the program is trying to do just after it looks for those .mo files.

[EnigmA]

Hello,
Well im New to SRCDS on Linux and i keep getting this error

Auto detecting CPU


Using default binary.
Server will auto-restart if there is a crash.
/root/srcds_1/srcds_run: line 344: 29083 Illegal instruction $HL_CMD
Add "-debug" to the /root/srcds_1/srcds_run command line to generate a debug.log to help with solving this problem
Wed Dec 31 12:24:57 EST 2008: Server restart in 10 seconds



If anyone can help me that would be Great

Thanks,
Anthony

[Oldboy]

Hi,

I`ve solved this problem by deleting es_tools from server. So, if you have installed it, try to delete and restart your server.