ai-forgot-these-cards Help

Troubleshooting

Port conflicts

  • If the backend can’t start, check whether APP_SERVER_PORT is already in use.

  • If llama.cpp runs on 8080, it will conflict with the backend; prefer 8087.

Can’t reach llama.cpp from the container

If llama.cpp is running on the host, the app container must be able to reach it.

  • Prefer SPRING_AI_OPENAI_CHAT_BASE_URL=http://host.docker.internal:<port>

  • Ensure your Docker setup supports the host gateway mapping

Nexus builds fail on first run

Nexus can take 1–2 minutes to become ready. If builds fail to reach the mirror, wait until Nexus is up and retry.

Database export/import issues

  • make export-db/make import-db require local Postgres client tools.

  • If you don’t have them, use make export-db-container and make import-db-container.

First place to look

  • App logs: make tail-tomcat-logs

Last modified: 16 December 2025