C++ Crash in monster combat?

[mano368]

can anyone help me understand what problem in the server functions are causing the crash?

Using Nekiro Downgrade 1.5 7.72

Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x000055fa5b6e8f94 in Monster::doAttacking(unsigned int) ()
[Current thread is 1 (Thread 0x7f8f3aa06700 (LWP 137994))]
(gdb) bt full
#0  0x000055fa5b6e8f94 in Monster::doAttacking(unsigned int) ()
No symbol table info available.
#1  0x000055fa5b52c76b in Creature::onAttacking(unsigned int) ()
No symbol table info available.
#2  0x000055fa5b56d0a4 in Game::checkCreatures(unsigned long) ()
No symbol table info available.
#3  0x000055fa5b59d4cd in void std::__invoke_impl<void, void (Game::*&)(unsigned long), Game*&, unsigned long&>(std::__invoke_memfun_deref, void (Game::*&)(unsigned long), Game*&, unsigned long&) ()
No symbol table info available.
#4  0x000055fa5b59894f in std::__invoke_result<void (Game::*&)(unsigned long), Game*&, unsigned long&>::type std::__invoke<void (Game::*&)(unsigned long), Game*&, unsigned long&>(void (Game::*&)(unsigned long), Game*&, unsigned long&) ()
No symbol table info available.
#5  0x000055fa5b59246c in void std::_Bind<void (Game::*(Game*, unsigned long))(unsigned long)>::__call<void, , 0ul, 1ul>(std::tuple<>&&, std::_Index_tuple<0ul, 1ul>) ()
No symbol table info available.
#6  0x000055fa5b589f45 in void std::_Bind<void (Game::*(Game*, unsigned long))(unsigned long)>::operator()<, void>() ()
No symbol table info available.
#7  0x000055fa5b582223 in std::_Function_handler<void (), std::_Bind<void (Game::*(Game*, unsigned long))(unsigned long)> >::_M_invoke(std::_Any_data const&) ()
No symbol table info available.
#8  0x000055fa5b7eaece in std::function<void ()>::operator()() const ()
No symbol table info available.
#9  0x000055fa5b7ea7e2 in Task::operator()() ()
No symbol table info available.
#10 0x000055fa5b7eaa9c in Dispatcher::threadMain() ()
No symbol table info available.
#11 0x000055fa5b72e96e in void std::__invoke_impl<void, void (Dispatcher::*)(), Dispatcher*>(std::__invoke_memfun_deref, void (Dispatcher::*&&)(), Dispatcher*&&) ()
No symbol table info available.
#12 0x000055fa5b72e703 in std::__invoke_result<void (Dispatcher::*)(), Dispatcher*>::type std::__invoke<void (Dispatcher::*)(), Dispatcher*>(void (Dispatcher::*&&)(), Dispatcher*&&) ()
No symbol table info available.
--Type <RET> for more, q to quit, c to continue without paging--

 

[Sarah Wesker]

Try checking if your sources have the changes from Combat_raw to Combat_ptr, that solved many problems with combat objects, besides fixing the issue with /reload spells and addEvent.

 

[mano368]

Thanks, I'll take a look at the ones you mentioned

 

[mano368]

Try checking if your sources have the changes from Combat_raw to Combat_ptr, that solved many problems with combat objects, besides fixing the issue with /reload spells and addEvent.

okay, a brief search in the sources of my server returned combat_prt, so I imagine that it already has this change, as for the change in the reload of spells and addEvent, I still haven't found any changes.

I remember that OT crashed when I put a Ferumbras to fight a Morgaroth XD, which seemed like just a joke returned the crash.

 

[clouf]

okay, a brief search in the sources of my server returned combat_prt, so I imagine that it already has this change, as for the change in the reload of spells and addEvent, I still haven't found any changes.

I remember that OT crashed when I put a Ferumbras to fight a Morgaroth XD, which seemed like just a joke returned the crash.

Meaby, these bosses' attacks are from the new Tibia and should not be used in lowered protocols

 

[Gesior.pl]

I remember that OT crashed when I put a Ferumbras to fight a Morgaroth XD, which seemed like just a joke returned the crash.

Compile engine in debug mode. Then we will get report with variables and line in which it crashed.
Did you run any /reload before crash?

 

[mano368]

Compile engine in debug mode. Then we will get report with variables and line in which it crashed.
Did you run any /reload before crash?

Yes, I reloaded the scripts almost an hour before the crash.
The same in another crash days ago.

Crash time 7:13pm, reload time 6:36pm

So, is it involved with reloading scripts? My spells are revscripts too, even monsters

 

[Gesior.pl]

So, is it involved with reloading scripts?

Some scripts reloads may result in delayed crash, when something that was wrongly reloaded tries to execute some script, that does not exist in RAM anymore. /reload is for development environment, not for real server with players online.

 

[Sarah Wesker]

okay, a brief search in the sources of my server returned combat_prt, so I imagine that it already has this change, as for the change in the reload of spells and addEvent, I still haven't found any changes.

I remember that OT crashed when I put a Ferumbras to fight a Morgaroth XD, which seemed like just a joke returned the crash.

Is this reproducible or just a one-time thing?

How often does it happen, have you already done tests?

 

[mano368]

Is this reproducible or just a one-time thing?

How often does it happen, have you already done tests?

probably yes, I just don't know how exactly, what I can say is, I used reload almost an hour before, then my friend summoned a morgaroth and a ferumbras to fight and in the middle of that battle the crash came

I did try after I saw the crashlog mentioning creatures, but I didn't know about the reload so I didn't even test it in the same way. I've had a crash before, also with reload, but I imagine that in these cases it was with normal monsters or something similar. I'll take a look here at the log and share the commands from moments before the crash.

In total I believe I had 3 crashes, but I couldn't get the crashlog of 2 because I had to do a configuration in Linux that I only found out about after reading a post by Gesior.pl, the only logs I have access to are from the console and the gms