mudler
port already in us, but I killed it?
LocalAI version: 3.6.0
Environment, CPU architecture, OS, and Version: Arch Linux Garuda Mokka KDE, RTX 3060, ryzen 7 2700X
Describe the bug it says adress already in use but I killed it before
To Reproduce sudo local-ai -run
Expected behavior it should work
Logs ❯ sudo ss -ltnp 'sport = :8080' State Recv-Q Send-Q Local Address:Port Peer Address:Port Process LISTEN 0 32768 0.0.0.0:8080 0.0.0.0:* users:(("local-ai",pid=29489,fd=6))
dervarex@dervarex in ~ as 🧙 ❯ sudo kill 29489
dervarex@dervarex in ~ as 🧙 ❯ sudo local-ai run 3:00PM INF Starting LocalAI using 16 threads, with models path: /usr/share/local-ai/models 3:00PM INF LocalAI version: v3.6.0 (8fb95686af9c295d68a3b7b5cb520dd2bf914bd4) 3:00PM WRN Failed to read system backends, proceeding with user-managed backends error="open /usr/share/localai/backends: no such file or directory" 3:00PM INF Capability automatically detected, set LOCALAI_FORCE_META_BACKEND_CAPABILITY to override Capability=nvidia 3:00PM WRN VRAM is less than 4GB, defaulting to CPU. Set LOCALAI_FORCE_META_BACKEND_CAPABILITY to override 3:00PM INF Preloading models from /usr/share/local-ai/models 3:00PM INF core/startup process completed! 3:00PM FTL Error running the application error="failed to listen: listen tcp4 0.0.0.0:8080: bind: address alreadyin use"
dervarex@dervarex in ~ as 🧙 [✖]
Additional context
