Skip to main content

LLMBoost

Issues with the Services → LLMBoost integration.

"Services → LLMBoost" page shows Disconnected​

The link is always in the sidebar. Disconnected means none of the registered agents responded on port 30080 (LLMBoost NodePort) or 6443 (K8s API).

  • Verify the LLMBoost cluster is running: lbh cluster status on the K8s host.
  • Confirm the agent's host IP can reach https://<k8s-ip>:6443/healthz (curl with -k).
  • If your K8s nodes are on a different subnet, set the endpoints explicitly via LLMBOOST_API_ENDPOINT / LLMBOOST_K8S_API_URL and restart the master.

LLMBoost page shows models but no deployments​

The K8s API needs a service-account token to list deployment CRDs. See Getting Started → First-Time Setup → LLMBoost for the two kubectl commands that create the read-only token.

Token expired​

If you provisioned the token with --duration=8760h (1 year), it will eventually expire and the deployment / pod views go blank while metrics keep working. Re-run the kubectl create token command and replace LLMBOOST_K8S_TOKEN on the master.

Endpoint mismatch (/api/ vs root)​

The official LLMBoost Helm chart serves the API at /api/ (e.g. http://<ip>:30080/api/v1/models). Standalone deployments serve at the root (e.g. http://<ip>:30080/v1/models). Set the endpoint accordingly:

mac env set LLMBOOST_API_ENDPOINT=http://<ip>:30080/api    # Helm chart
# OR
mac env set LLMBOOST_API_ENDPOINT=http://<ip>:30080        # standalone

Collecting a Support Bundle​

When opening a ticket, please include the output of:

mac status
docker logs mac-api-server --tail 500
docker logs mac-prometheus --tail 200
curl -s http://localhost:29090/api/v1/targets
uname -a
dpkg -l mango-ai-center

Send it to contact@mangoboost.io along with a description of what you expected and what you observed.