spektr
2f03a7e624
10.4-Tests bestanden: Idempotenz (changed=0), kontrollierter Neustart erhaelt Profil/Modell, bewusster Stopp bleibt Stopp, Profilwechsel in beide Richtungen ueber die Rolle fehlerfrei ohne erneuten Download. Befundkorrektur 5.3: direkter systemctl start der zweiten Unit fuehrt KEINEN geordneten Wechsel aus - beide Modelle laden in der UEbergangsphase parallel (beobachtet 01.11.2026). Conflicts= sichert nur den Endzustand; Before= aus den Quadlets entfernt. Der geordnete Stop-vor-Start-Wechsel ist ausschliesslich die Rollen-Umschaltoperation. Host-Reboot-Test an den Betreiber uebertragen. Generated by Mistral Vibe. Co-Authored-By: Mistral Vibe <vibe@mistral.ai>
LLM serving stack – Ansible control project
Executes the LLM serving stack for exactly one environment on
ghostnet-ki (GNAI-008 section 7.4). Host baseline, Podman runtime,
workload accounts and AMD GPU access are managed by
../../host/ansible; this project never
changes host state.
Layout
inventories/<environment>/– one inventory per KI environment (currentlytesting); deployment uses theansible-deployaccount withbecome: true.playbooks/ai.yml– appliesmodel_storetheninference_engine.roles/model_store/– shared, pinned model weights under/var/lib/ghostnet-ai/shared/models(GNAI-010 sections 3 and 4).roles/inference_engine/– rootless Quadlets, network, image provisioning and the exclusive profile switch for llama.cpp (GNAI-010 sections 5 and 6).
Control node setup
Use the repository-wide control venv built from
../../../shared/ansible/requirements-control.txt
(ansible-core==2.21.4), for example:
python3 -m venv .venv
.venv/bin/pip install -r shared/ansible/requirements-control.txt
cd llm-server/llm/ansible
../../../.venv/bin/ansible-playbook -i inventories/testing playbooks/ai.yml --syntax-check
Usage (Testing)
All commands use the operator deploy key
~/.ssh/id_ed25519_ghostnet_ai_ansible_deploy (see AGENTS.md; no ssh-agent
required):
# Preview only – no changes, no downloads, no service start:
../../../.venv/bin/ansible-playbook -i inventories/testing playbooks/ai.yml \
--check --diff --private-key ~/.ssh/id_ed25519_ghostnet_ai_ansible_deploy
# Provision model weights only (tag models):
../../../.venv/bin/ansible-playbook -i inventories/testing playbooks/ai.yml \
--tags models --private-key ~/.ssh/id_ed25519_ghostnet_ai_ansible_deploy
# Render quadlets and start/verify the selected profile (tag service):
../../../.venv/bin/ansible-playbook -i inventories/testing playbooks/ai.yml \
--tags service --private-key ~/.ssh/id_ed25519_ghostnet_ai_ansible_deploy
inference_engine_active_profile selects which rendered quadlet is started;
all other profiles stay rendered but stopped. Check mode downloads no model
and starts no service. A second identical run must report changed=0
(GNAI-010 section 10.4).
The full technical acceptance (64k/128k start, VRAM distribution, prefill ladder, restart behaviour) is an operator-run protocol documented in GNAI-010; this role automates deployment and basic health only.