{"id":"7cvBP9FxE9F8_hDAaC8r8","title":"text code snippet (12 lines)","language":"text","tags":[],"content":"# 1. on .35 — is the llama server even up, and how is the port bound?\ndocker ps --format '{{.Names}}\\t{{.Ports}}' | grep -i llama\n# want 0.0.0.0:8080->8080, NOT 127.0.0.1:8080->8080\n\n# 2. did it crash or is it mid-load?\ndocker logs --tail 50 <llama-container>\n\n# 3. from a different box (not .35 itself):\ncurl -s http://192.168.50.35:8080/v1/models\n\n# 4. from inside the hindsight container, exactly what triage does:\ndocker exec -it hindsight-api curl -s http://192.168.50.35:8080/v1/models\n","created_at":"2026-11-13T19:51:42.365Z"}