Jump to content

Loopy

Members
  • Posts

    3
  • Joined

  • Last visited

  • Donations

    0.00 GBP 

Core Infomation

  • Core
    Zero

Loopy's Achievements

Newbie

Newbie (1/3)

0

Reputation

  1. Fixed mobs using Backstab while frontally facing the player. fix_frontal_backstab.diff
  2. This bug is still around. The leveling is very challenging as a result. Mobs are able to backstab while front facing, and they're doing it on cooldown for the most part.
  3. I've successfully built out the Zero server and everything appears to be running smoothly. However, after doing some monitoring, it appears that there is a slow but constant memory leak going on if random AI Bots are enabled. Details: Platform: Ubuntu VM, 8gb ram, 2 core CPU. Core: Zero Configuration: Everything more or less at default, Random AI bots enabled, 50 minimum - 100 maximum, deleterandombotaccounts = 1 What happens: Memory utilization starts at around 8% (as per monitoring via 'top'), and steadily rises up to roughly 22%+ after about 2hrs, continuing to rise. I am not a coding expert, however i did use chatGPT to give me some pointers on where to look for the sources of memory leakage. I used valgrind to identify which functions were improperly handling memory allocation, and after running the tool for about 45 mins, these are the functions it identified: Memory Allocation Location: The memory is allocated in the operator new[] function (which is used for dynamic arrays), as seen in the trace. The leak is most prominent in ai::NextAction::clone, which suggests that each time this action is cloned, memory is allocated but never freed. Leak Propagation: The leak then propagates through the following functions: ai::Engine::ProcessTriggers() ai::Engine::DoNextAction() PlayerbotAI::DoNextAction() PlayerbotAIBase::UpdateAI() Player::Update() Map::Update() MapUpdateRequest::call() Other functions involved in managing the AI updates in the game world I've experimented by turning off random AI bots. Memory utilization started at 4.8% at startup, and after about an hour held steady at pretty much the same level. Any ideas on how to remedy this issue from my end, if possible? Are there others experiencing the same issue, or is there something about my specific deployment that is causing this?
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. Privacy Policy Terms of Use