Xenithar

Alright, Foereaper explained changes which do not appear to be documented on our forum to me. The addition of the new cmake parameter will clear PART of this up. We now need a parameter for where to place binaries. I will suggest that when these parameters are omitted, we default to the correct locations. Under Unix/Linux/BSD/Apple/Solaris/etc this will be "etc" for configuration files and "bin" for the binaries. Under Windows I would suggest "%Userdocs%/My Games/MaNGOS" for the configuration files and wherever the user wants for the binaries. This way when these parameters are omitted we still get a sane solution on all systems.

I backed up my build before building the current build and after restoring it, I cannot reproduce this bug on R20. In fact it has not happened to any of us since R20 came out. Maybe something else is triggering it?

What was broken in the old system? It worked fine on several platforms for me. This, from my perspective, looks like it is being made to be more like a Windows application.

Alright, I will try the internal ACE library once the proper file locations have been restored and it will run under anything but Windows. For now, I am stuck.

Why are we changing build systems AGAIN? This is why we keep having issues. I am unable to even start the binaries for debugging.

I see. The default location for binaries and for configuration files is not sane in any environment I use. Also, we need ACE working. I have had all kinds of trouble with the internal ACE in the past, external works. This should be top priority. *EDIT* Let's handle the install locations in that ticket. This one is still about no longer being able to use ACE.

It compiled, but there is a new problem. The binaries did not get installed to /bin, they installed in the root along with the configuration files. In other words, I did not get /bin or /etc as it should. This would break installing it on any OS as an OS package aside from possibly Windows. *UPDATE* Upon creating the correct directories and sorting the files into it, it cannot find the configuration files. Something is VERY broken here outside of not being able to use the external libraries.

I will post the result in about 30min.

Same as always, Debian GNU/Linux. Parameters are as follows. [code] -DDEBUG=1 -DUSE_STD_MALLOC=1 -DACE_USE_EXTERNAL=1 -DPOSTGRESQL=0 -DBUILD_TOOLS=0 -DSCRIPT_LIB_ELUNA=1 -DSCRIPT_LIB_SD2=1 -DCMAKE_INSTALL_PREFIX=/home/???/zero [/code] The question marks simply hide my username. I don't want it shown.

I use external ACE, which is present. Nothing changed. I compiled, went on vacation, returned, cloned fresh, compile failed. I have not changed my system in any way. [code] ???@WoW01:~$ dpkg --get-selections | grep ace libace-6.0.3 install libace-dev install traceroute install ???@WoW01:~$ [/code] So it is there. Something changed since I left two weeks ago. *EDIT* Also, this: [code] ???@WoW01:~/zero/src/server/build$ make [ 0%] Built target genrev [ 0%] Built target revision.h [ 1%] Built target detour [ 3%] Built target recast [ 11%] Built target g3dlite [ 14%] Built target zlib [ 15%] Built target bzip2 [ 16%] Built target libmpq [ 23%] Built target lualib [ 24%] Built target framework [ 29%] Built target shared [ 30%] Built target LuaEngine [ 30%] Built target pchdef.cpp_dephelp [ 30%] Built target generate_pchdef.cpp [ 30%] Built target precompiled.cpp_dephelp [ 31%] Built target generate_precompiled.cpp [ 68%] Built target mangosscript [ 98%] Built target game Linking CXX executable realmd /usr/bin/ld: cannot find -lace collect2: error: ld returned 1 exit status make[2]: *** [src/realmd/realmd] Error 1 make[1]: *** [src/realmd/CMakeFiles/realmd.dir/all] Error 2 make: *** [all] Error 2 [/code] *UPDATE* The "link.txt" file for realmd is all screwed up. [code] /usr/bin/c++ -g -g -Wall -Wfatal-errors -Wextra -Winvalid-pch -Wl,--no-as-needed -ldl -pthread -lrt CMakeFiles/realmd.dir/AuthSocket.cpp.o CMakeFiles/realmd.dir/BufferedSocket.cpp.o CMakeFiles/realmd.dir/Main.cpp.o CMakeFiles/realmd.dir/PatchHandler.cpp.o CMakeFiles/realmd.dir/RealmList.cpp.o -o realmd -rdynamic ../shared/libshared.a ../framework/libframework.a -lace -lmysqlclient_r -lssl -lcrypto -Wl,-rpath,/home/???/zero/lib [/code] First, there are two "-g" parameters. Why? Next, I see "-Wl" with comma, but no commas for the other parameters. If we're passing multiple parameters to the linker, it should be a comma-separated list. Finally, I see additional parameters at the end of the command and the another "-Wl" with a proper set of commands. It does not look correct to me. Can somebody verify that we're supposed to have it this way? I do not believe we should have to "-g" parameters and I am still not following the entire syntax here.

I have a ticket on the bot being broken already. Exactly as Max says. Starts fine, but nothing is shown in-game and nothing is ever bought from players.

I looked and only saw ONE commit that appeared to be for R20, and it changed variable (constant?) names. The others are before I started having issues. The only recent one is linked below, but I do not see how this would cause the program to not work at all. [url]https://github.com/mangoszero/server/commit/b1219090a5b5f222f26417d4dc62f64c604d0453[/url]

And another invalid node overlap. Same as the last two.

Another one. I am not going to paste this because it is the same thing. I got "invalid node overlap" again.

I was under the impression AHBot was not modified. This changes things. I will start with those. Thanks. I am also looking at the dang crash that happens all the time, but it appears to be with vmaps and I do not understand vmaps.

Alright, this time I have no clue why the crash happened. Mom was playing a new warlock, killed the required mobs and hearthed to return to the starting spot. When she got there, nothing spawned or moved. [code] WORLD: Sent SMSG_QUESTGIVER_STATUS for Creature (Entry: 3145 Guid: 3444) WORLD: Received opcode CMSG_QUESTGIVER_STATUS_QUERY - for Player Durqas (Guid: 31) to Creature (Entry: 5887 Guid: 3417) WORLD: Sent SMSG_QUESTGIVER_STATUS for Creature (Entry: 5887 Guid: 3417) WORLD: Received opcode CMSG_QUESTGIVER_STATUS_QUERY - for Player Durqas (Guid: 31) to Creature (Entry: 3143 Guid: 3443) WORLD: Sent SMSG_QUESTGIVER_STATUS for Creature (Entry: 3143 Guid: 3443) WORLD: Received opcode CMSG_QUESTGIVER_STATUS_QUERY - for Player Durqas (Guid: 31) to Creature (Entry: 3882 Guid: 7660) WORLD: Sent SMSG_QUESTGIVER_STATUS for Creature (Entry: 3882 Guid: 7660) WORLD: Received opcode CMSG_QUESTGIVER_STATUS_QUERY - for Player Durqas (Guid: 31) to Creature (Entry: 3158 Guid: 7282) WORLD: Sent SMSG_QUESTGIVER_STATUS for Creature (Entry: 3158 Guid: 7282) WORLD: Received opcode CMSG_QUESTGIVER_STATUS_QUERY - for Player Durqas (Guid: 31) to Creature (Entry: 9796 Guid: 704) WORLD: Sent SMSG_QUESTGIVER_STATUS for Creature (Entry: 9796 Guid: 704) terminate called after throwing an instance of 'std::logic_error' what(): invalid node overlap Program received signal SIGABRT, Aborted. [Switching to Thread 0xb20f3b70 (LWP 3063)] 0xb7fe1424 in __kernel_vsyscall () (gdb) backtrace #0 0xb7fe1424 in __kernel_vsyscall () #1 0xb7697681 in raise () from /lib/i386-linux-gnu/i686/cmov/libc.so.6 #2 0xb769aab2 in abort () from /lib/i386-linux-gnu/i686/cmov/libc.so.6 #3 0xb787e28d in __gnu_cxx::__verbose_terminate_handler() () from /usr/lib/i386-linux-gnu/libstdc++.so.6 #4 0xb787bef3 in ?? () from /usr/lib/i386-linux-gnu/libstdc++.so.6 #5 0xb787bf2f in std::terminate() () from /usr/lib/i386-linux-gnu/libstdc++.so.6 #6 0xb787c1ce in __cxa_throw () from /usr/lib/i386-linux-gnu/libstdc++.so.6 #7 0x089389d7 in BIH::subdivide (this=0xcac10d0, left=31, right=42, tempTree=..., dat=..., gridBox=..., nodeBox=..., nodeIndex=102, depth=9, stats=...) at /home/---/zero/src/server/src/game/vmap/BIH.cpp:68 #8 0x0893955b in BIH::subdivide (this=0xcac10d0, left=23, right=30, tempTree=..., dat=..., gridBox=..., nodeBox=..., nodeIndex=96, depth=8, stats=...) at /home/---/zero/src/server/src/game/vmap/BIH.cpp:252 #9 0x08938e48 in BIH::subdivide (this=0xcac10d0, left=23, right=30, tempTree=..., dat=..., gridBox=..., nodeBox=..., nodeIndex=90, depth=7, stats=...) at /home/---/zero/src/server/src/game/vmap/BIH.cpp:128 #10 0x089394dd in BIH::subdivide (this=0xcac10d0, left=23, right=42, tempTree=..., dat=..., gridBox=..., nodeBox=..., nodeIndex=24, depth=6, stats=...) at /home/---/zero/src/server/src/game/vmap/BIH.cpp:248 #11 0x0893955b in BIH::subdivide (this=0xcac10d0, left=0, right=22, tempTree=..., dat=..., gridBox=..., nodeBox=..., nodeIndex=18, depth=5, stats=...) at /home/---/zero/src/server/src/game/vmap/BIH.cpp:252 #12 0x089394dd in BIH::subdivide (this=0xcac10d0, left=0, right=48, tempTree=..., dat=..., gridBox=..., nodeBox=..., nodeIndex=9, depth=3, stats=...) at /home/---/zero/src/server/src/game/vmap/BIH.cpp:248 #13 0x089394dd in BIH::subdivide (this=0xcac10d0, left=0, right=49, tempTree=..., dat=..., gridBox=..., nodeBox=..., nodeIndex=0, depth=1, stats=...) at /home/---/zero/src/server/src/game/vmap/BIH.cpp:248 #14 0x08938721 in BIH::buildHierarchy (this=0xcac10d0, tempTree=..., dat=..., stats=...) at /home/---/zero/src/server/src/game/vmap/BIH.cpp:38 #15 0x0893c0fb in BIH::build >(G3D::Array const&, void (&)(GameObjectModel const*, G3D::AABox&), unsigned int, bool) ( this=0xcac10d0, primitives=..., getBounds= @0x893a5cd: {void (const GameObjectModel *, G3D::AABox &)} 0x893a5cd ::getBounds2(GameObjectModel const*, G3D::AABox&)>, leafSize=3, printStats=false) at /home/---/zero/src/server/src/game/vmap/BIH.h:156 #16 0x0893b223 in BIHWrap >::balance (this=0xcac10d0) at /home/---/zero/src/server/src/game/vmap/BIHWrap.h:147 #17 0x0893a9eb in RegularGrid2D >, NodeCreator > >, PositionTrait >::balance (this=0xb30bd68) at /home/---/zero/src/server/src/game/vmap/RegularGrid.h:148 #18 0x0893a69b in DynTreeImpl::balance (this=0xb30bd68) at /home/---/zero/src/server/src/game/vmap/DynamicTree.cpp:86 #19 0x0893a722 in DynTreeImpl::update (this=0xb30bd68, difftime=338) at /home/---/zero/src/server/src/game/vmap/DynamicTree.cpp:100 #20 0x08939e98 in DynamicMapTree::update (this=0xb9acf7c, t_diff=338) at /home/---/zero/src/server/src/game/vmap/DynamicTree.cpp:144 #21 0x08842624 in Map::Update (this=0xb987de0, t_diff=@0xb20f323c: 338) at /home/---/zero/src/server/src/game/Map.cpp:460 #22 0x08857f52 in MapManager::Update (this=0xb2ff358, diff=338) at /home/---/zero/src/server/src/game/MapManager.cpp:184 #23 0x088e8401 in World::Update (this=0x8ce9348, diff=338) at /home/---/zero/src/server/src/game/World.cpp:1498 #24 0x0855402f in WorldRunnable::run (this=0xa4f2b70) at /home/---/zero/src/server/src/mangosd/WorldRunnable.cpp:66 #25 0x0897dca5 in ACE_Based::Thread::ThreadTask (param=0xa4f2b70) at /home/---/zero/src/server/src/shared/Threading.cpp:197 #26 0xb7f50171 in ACE_OS_Thread_Adapter::invoke() () from /usr/lib/libACE-6.0.3.so #27 0xb7f121cf in ace_thread_adapter () from /usr/lib/libACE-6.0.3.so #28 0xb77d6c39 in start_thread () from /lib/i386-linux-gnu/i686/cmov/libpthread.so.0 #29 0xb7743a6e in clone () from /lib/i386-linux-gnu/i686/cmov/libc.so.6 [/code] Another "invalid node overlap". What "nodes" are throwing these errors? Also, I built the server from a fresh clone this morning, so it should be up to date.

Yeah but this is Linux, on a console, same way it has run for years. I don't start it with a script or anything. I just know it worked in R19 and does not work in R20, so I believe it is an issue in the code somewhere, but I need a starting point to even begin looking.

Alright, this is getting bad. It has become VERY difficult to (without using GM commands) keep cash for skills when levelling. What caused the AHBot to break in R20? I am assuming nobody changed the bot itself, so something else must have changed to cause the break. What was changed that would break the bot? I need to look into this.

I will be honest here, I have NEVER had MaNGOS freeze on my servers. I have had crashes from bugs, but never from a hang. I suppose this would be good if it was to inform the server owner that the process hung and did not crash due to some DB or coding error.

Another. This time it was "negative node extents". We had just finished a DM run and I teleported us to Tarren Mill. We were selling things to vendors and BAM, it died. [code] WORLD: Sent SMSG_QUESTGIVER_STATUS for Creature (Entry: 2393 Guid: 15292) WORLD: Received opcode CMSG_QUESTGIVER_STATUS_QUERY - for Player Marrek (Guid: 3) to Creature (Entry: 2216 Guid: 15301) WORLD: Sent SMSG_QUESTGIVER_STATUS for Creature (Entry: 2216 Guid: 15301) WORLD: Received opcode CMSG_QUESTGIVER_STATUS_QUERY - for Player Marrek (Guid: 3) to Creature (Entry: 2389 Guid: 15299) WORLD: Sent SMSG_QUESTGIVER_STATUS for Creature (Entry: 2389 Guid: 15299) Aura 22 now is remove mode 5 Aura 42 now is remove mode 5 Aura 22 now is remove mode 5 Aura 42 now is remove mode 5 Aura 22 now is remove mode 5 Aura 42 now is remove mode 5 Aura 42 now is remove mode 5 Aura 13 now is remove mode 5 Aura 138 now is remove mode 5 terminate called after throwing an instance of 'std::logic_error' what(): negative node extents Program received signal SIGABRT, Aborted. [Switching to Thread 0xb20f3b70 (LWP 2934)] 0xb7fe1424 in __kernel_vsyscall () (gdb) backtrace #0 0xb7fe1424 in __kernel_vsyscall () #1 0xb7697681 in raise () from /lib/i386-linux-gnu/i686/cmov/libc.so.6 #2 0xb769aab2 in abort () from /lib/i386-linux-gnu/i686/cmov/libc.so.6 #3 0xb787e28d in __gnu_cxx::__verbose_terminate_handler() () from /usr/lib/i386-linux-gnu/libstdc++.so.6 #4 0xb787bef3 in ?? () from /usr/lib/i386-linux-gnu/libstdc++.so.6 #5 0xb787bf2f in std::terminate() () from /usr/lib/i386-linux-gnu/libstdc++.so.6 #6 0xb787c1ce in __cxa_throw () from /usr/lib/i386-linux-gnu/libstdc++.so.6 #7 0x089388b7 in BIH::subdivide (this=0xd2f58f8, left=0, right=3, tempTree=..., dat=..., gridBox=..., nodeBox=..., nodeIndex=0, depth=1, stats=...) at /home/---/zero/src/server/src/game/vmap/BIH.cpp:62 #8 0x08938721 in BIH::buildHierarchy (this=0xd2f58f8, tempTree=..., dat=..., stats=...) at /home/---/zero/src/server/src/game/vmap/BIH.cpp:38 #9 0x0893c0fb in BIH::build >(G3D::Array const&, void (&)(GameObjectModel const*, G3D::AABox&), unsigned int, bool) ( this=0xd2f58f8, primitives=..., getBounds= @0x893a5cd: {void (const GameObjectModel *, G3D::AABox &)} 0x893a5cd ::getBounds2(GameObjectModel const*, G3D::AABox&)>, leafSize=3, printStats=false) at /home/---/zero/src/server/src/game/vmap/BIH.h:156 #10 0x0893b223 in BIHWrap >::balan ce (this=0xd2f58f8) at /home/---/zero/src/server/src/game/vmap/BIHWrap.h:147 #11 0x0893a9eb in RegularGrid2D >, NodeCreator > >, PositionTrait >::balance (this=0xdd42018) at /home/---/zero/src/server/src/game/vmap/RegularGrid.h:148 #12 0x0893a69b in DynTreeImpl::balance (this=0xdd42018) at /home/---/zero/src/server/src/game/vmap/DynamicTree.cpp:86 #13 0x0893a722 in DynTreeImpl::update (this=0xdd42018, difftime=150) at /home/---/zero/src/server/src/game/vmap/DynamicTree.cpp:100 #14 0x08939e98 in DynamicMapTree::update (this=0xd8034a4, t_diff=150) at /home/---/zero/src/server/src/game/vmap/DynamicTree.cpp:144 #15 0x08842624 in Map::Update (this=0xd7de308, t_diff=@0xb20f323c: 150) at /home/---/zero/src/server/src/game/Map.cpp:460 #16 0x088467bc in DungeonMap::Update (this=0xd7de308, t_diff=@0xb20f323c: 150) at /home/---/zero/src/server/src/game/Map.cpp:1384 #17 0x08857f52 in MapManager::Update (this=0xb2fed30, diff=51) at /home/---/zero/src/server/src/game/MapManager.cpp:184 #18 0x088e8401 in World::Update (this=0x8ce92b0, diff=51) at /home/---/zero/src/server/src/game/World.cpp:1498 #19 0x0855402f in WorldRunnable::run (this=0xa4e1ef0) at /home/---/zero/src/server/src/mangosd/WorldRunnable.cpp:66 #20 0x0897dca5 in ACE_Based::Thread::ThreadTask (param=0xa4e1ef0) at /home/---/zero/src/server/src/shared/Threading.cpp:197 #21 0xb7f50171 in ACE_OS_Thread_Adapter::invoke() () from /usr/lib/libACE-6.0.3.so #22 0xb7f121cf in ace_thread_adapter () from /usr/lib/libACE-6.0.3.so #23 0xb77d6c39 in start_thread () from /lib/i386-linux-gnu/i686/cmov/libpthread.so.0 #24 0xb7743a6e in clone () from /lib/i386-linux-gnu/i686/cmov/libc.so.6 [/code] This is kind of out of hand. What nodes are crashing the server, ores?