vllm-cpu-release Container Remains Running after test completes

[jharriga]

DUT system continues to run vllm container after test completes.

ansible-playbook -v -i inventory/hosts.yml llm-benchmark-auto.yml \ -e "test_model= TinyLlama/TinyLlama-1.1B-Chat-v1.0" \ -e "workload_type=chat" \ -e "requested_cores=16"

Final Ansible messages:

PLAY [Auto-Configured LLM Test - Optional Cleanup] *****************************

TASK [Gathering Facts] *********************************************************
ok: [vllm-server]

TASK [Stop and remove vLLM container] ******************************************
skipping: [vllm-server] => {"changed": false, "false_condition": "cleanup_after_test | default(false) | bool", "skip_reason": "Conditional result was False"}

TASK [Display cleanup status] **************************************************
skipping: [vllm-server] => {"false_condition": "cleanup_after_test | default(false) | bool"}

PLAY RECAP *********************************************************************
guidellm-client : ok=41 changed=7 unreachable=0 failed=0 skipped=22 rescued=0 ignored=0
localhost : ok=8 changed=0 unreachable=0 failed=0 skipped=4 rescued=0 ignored=0
vllm-server : ok=76 changed=6 unreachable=0 failed=0 skipped=22 rescued=0 ignored=0

dut# podman ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES e541b9771546 docker.io/vllm/vllm-openai-cpu:v0.18.0 --model --host 0.... About an hour ago Up About an hour vllm-server