I've recently encountered an issue on the latest build of TFC (version : 48/1.1.2.2/Stdio 8308 secure (20)) whereby infecting an enemy guarantees a segfault within a few seconds. Here is a dump from the debug.log showing where the problem is:
Quote:
----------------------------------------------
CRASH: Tue Mar 3 20:07:32 CST 2020
Start Line: ./hlds_linux -game tfc +ip <removed> +maxplayers 17 +hpk_maxsize 1 -debug -console -noaff -pingboost 2 -norestart -pidfile hlds.27777.pid
[New LWP 27785]
[New LWP 27795]
[New LWP 27791]
[New LWP 27796]
[New LWP 27797]
[New LWP 27788]
[New LWP 27798]
[New LWP 27789]
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
Core was generated by `./hlds_linux -game tfc +ip <removed> +maxplayers 17 +hpk_maxsize 1 -debug -'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0 0xf2da0483 in Client_Damage(void*) () from tfc/addons/amxmodx/modules/tfcx_amxx_i386.so
[Current thread is 1 (Thread 0xf7bcf700 (LWP 27785))]
#0 0xf2da0483 in Client_Damage(void*) () from tfc/addons/amxmodx/modules/tfcx_amxx_i386.so
#1 0xf7a7e740 in userfilters () from /servers/akw/engine_i486.so
#2 0xf32e9010 in ?? ()
Backtrace stopped: previous frame inner to this frame (corrupt stack?)
No symbol table info available.
From To Syms Read Shared Object Library
0xf7f0e914 0xf7f54c78 Yes ./libstdc++.so.6
0xf7eb4ab0 0xf7eb5c64 Yes (*) /lib32/libdl.so.2
0xf7e989b0 0xf7ea9d1f Yes (*) /lib32/libpthread.so.0
0xf7cc95a0 0xf7e21e06 Yes (*) /lib32/libc.so.6
0xf7bef570 0xf7c78e95 Yes (*) /lib32/libm.so.6
0xf7f9da40 0xf7fb852b Yes (*) /lib/ld-linux.so.2
0xf7bd2e04 0xf7be2490 Yes ./libgcc_s.so.1
0xf76b35e0 0xf773970c Yes /servers/akw/engine_i486.so
0xf76585c0 0xf7662d74 Yes (*) ./libsteam_api.so
0xf764d8e0 0xf76517d4 Yes (*) /lib32/librt.so.1
0xf7634aa0 0xf7641428 Yes /servers/akw/filesystem_stdio.so
0xf5dcb0c0 0xf6f76d64 Yes (*) ./steamclient.so
0xf5c44670 0xf5cba020 Yes (*) ./crashhandler.so
0xf337e560 0xf3417128 Yes (*) /servers/akw/./tfc/addons/metamod/dlls/metamodn.so
0xf31baa10 0xf3285a8c Yes /servers/akw/tfc/dlls/tfc.so
0xf2f4bbc0 0xf2faddc7 Yes (*) /servers/akw/tfc/addons/amxmodx/dlls/amxmodx_mm_i386.so
0xf2d9b650 0xf2da1f38 Yes (*) tfc/addons/amxmodx/modules/tfcx_amxx_i386.so
0xf2471b60 0xf2478af4 Yes (*) /lib32/libnss_files.so.2
0xf2469b50 0xf246caf4 Yes (*) /lib32/libnss_dns.so.2
0xf2453640 0xf2460054 Yes (*) /lib32/libresolv.so.2
(*): Shared library is missing debugging information.
Stack level 0, frame at 0xff8cfca0:
eip = 0xf2da0483 in Client_Damage(void*); saved eip = 0xf7a7e740
called by frame at 0xff8cfca4
Arglist at 0xff8cfc98, args:
Locals at 0xff8cfc98, Previous frame's sp is 0xff8cfca0
Saved registers:
ebx at 0xff8cfc94, esi at 0xff8cfc8c, edi at 0xff8cfc90, eip at 0xff8cfc9c
End of crash report
----------------------------------------------
|
We've tested on AMX-1.9.0-5263, as well as a few 1.8.3 versions that we've used in the past: 5116, 5134, 5154. There were no issues on older builds of TFC as recent as (version : 48/1.1.2.2/Stdio 8196 secure (20)). As a temporary solution I would roll back the library myself but currently our server provider is standing in the way of that process.
We've also tried switching from Metamod 1.21-am to Metamod-P 1.21p37 to no avail.
I should note that this has been tested on a completely fresh server installation of TFC, metamod, and AMXX without any custom plugins present.
I would report this on the Alliedmodders bugzilla but I can't seem to find a way to select AMX Mod X as the product in question so I will try here.