From 96d8375663dc25ead236352c59ec1a04be024749 Mon Sep 17 00:00:00 2001
From: Yuan Tang <terrytangyuan@gmail.com>
Date: Fri, 3 Jan 2025 11:47:10 -0600
Subject: [PATCH] Fix incorrect entrypoint for broken `llama stack run` (#706)

This fixes the issue when using `llama stack run` by correctly
specifying entrypoint:

```
LLAMA_STACK_DIR=. llama stack run /home/yutang/.llama/distributions/llamastack-vllm/vllm-run.yaml
Using config file: /home/yutang/.llama/distributions/llamastack-vllm/vllm-run.yaml
+ command -v selinuxenabled
+ selinuxenabled
+ DOCKER_OPTS=' --security-opt label=disable'
+ mounts=
+ '[' -n . ']'
++ readlink -f .
+ mounts=' -v /home/yutang/repos/llama-stack:/app/llama-stack-source'
+ '[' -n '' ']'
+ version_tag=latest
+ '[' -n '' ']'
+ '[' -n . ']'
+ version_tag=dev
+ podman run --security-opt label=disable -it -p 5000:5000 -v /home/yutang/.llama/distributions/llamastack-vllm/vllm-run.yaml:/app/config.yaml -v /home/yutang/repos/llama-stack:/app/llama-stack-source localhost/distribution-vllm:dev python -m llama_stack.distribution.server.server --yaml-config /app/config.yaml --port 5000
usage: server.py
       [-h]
       [--yaml-config YAML_CONFIG]
       [--template TEMPLATE]
       [--port PORT]
       [--disable-ipv6]
       [--env ENV]
server.py: error: unrecognized arguments: python -m llama_stack.distribution.server.server
++ error_handler 88
++ echo 'Error occurred in script at line: 88'
Error occurred in script at line: 88
++ exit 1

```

---------

Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
---
 llama_stack/distribution/server/server.py   | 7 ++++++-
 llama_stack/distribution/start_container.sh | 7 +++----
 2 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/llama_stack/distribution/server/server.py b/llama_stack/distribution/server/server.py
index e432cca4e..8c1e41dc0 100644
--- a/llama_stack/distribution/server/server.py
+++ b/llama_stack/distribution/server/server.py
@@ -239,7 +239,12 @@ def main():
         "--template",
         help="One of the template names in llama_stack/templates (e.g., tgi, fireworks, remote-vllm, etc.)",
     )
-    parser.add_argument("--port", type=int, default=5000, help="Port to listen on")
+    parser.add_argument(
+        "--port",
+        type=int,
+        default=int(os.getenv("LLAMASTACK_PORT", 5000)),
+        help="Port to listen on",
+    )
     parser.add_argument(
         "--disable-ipv6", action="store_true", help="Whether to disable IPv6 support"
     )
diff --git a/llama_stack/distribution/start_container.sh b/llama_stack/distribution/start_container.sh
index 34476c8e0..3b7b55b97 100755
--- a/llama_stack/distribution/start_container.sh
+++ b/llama_stack/distribution/start_container.sh
@@ -90,7 +90,6 @@ $DOCKER_BINARY run $DOCKER_OPTS -it \
   $env_vars \
   -v "$yaml_config:/app/config.yaml" \
   $mounts \
-  $docker_image:$version_tag \
-  python -m llama_stack.distribution.server.server \
-  --yaml-config /app/config.yaml \
-  --port "$port"
+  --env LLAMASTACK_PORT=$port \
+  --entrypoint='["python", "-m", "llama_stack.distribution.server.server", "--yaml-config", "/app/config.yaml"]' \
+  $docker_image:$version_tag