Merge pull request #59 from stackhpc/feat/structlog

Switch to structlog for web app logging

.gitignore

@@ -5,6 +5,7 @@ __pycache__/
 **/*.secret
 .DS_Store
 .tox/
+**/.ruff_cache
 
 # Ignore local dev helpers
 test-values.y[a]ml

charts/azimuth-llm/values.schema.json

@@ -118,9 +118,7 @@
                         }
                     },
                     "required": [
-                        "model_name",
-                        "model_instruction"
-                    ]
+                        "model_name"                    ]
                 }
             }
         }

charts/azimuth-llm/values.yaml

@@ -92,8 +92,6 @@ ui:
   # available configuration options.
   appSettings:
     model_name: *model-name
-    model_instruction: |
-      You are a helpful AI assistant. Please respond appropriately.
     # Use local system fonts by default to avoid GDPR issues
     # with Gradio's defaults fonts which require fetching from
     # the Google fonts API. To restore default Gradio theme

web-apps/chat/app.py

@@ -1,18 +1,14 @@
-import logging
 import openai
-
+import utils
 import gradio as gr
 
 from urllib.parse import urljoin
 from langchain.schema import HumanMessage, AIMessage, SystemMessage
 from langchain_openai import ChatOpenAI
-from typing import Dict, List
+from typing import Dict
 from pydantic import BaseModel, ConfigDict
-from utils import LLMParams, load_settings
 
-logging.basicConfig()
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.INFO)
+log = utils.get_logger()
 
 
 class AppSettings(BaseModel):
@@ -22,7 +18,7 @@ class AppSettings(BaseModel):
     model_name: str
     model_instruction: str
     page_title: str
-    llm_params: LLMParams
+    llm_params: utils.LLMParams
     # Theme customisation
     theme_params: Dict[str, str | list]
     theme_params_extended: Dict[str, str]
@@ -32,8 +28,8 @@ class AppSettings(BaseModel):
     model_config = ConfigDict(protected_namespaces=(), extra="forbid")
 
 
-settings = AppSettings(**load_settings())
-logger.info(settings)
+settings = AppSettings(**utils.load_settings())
+log.info(settings)
 
 backend_url = str(settings.backend_url)
 backend_health_endpoint = urljoin(backend_url, "/health")
@@ -82,7 +78,7 @@ def inference(latest_message, history):
             context.append(HumanMessage(content=human))
             context.append(AIMessage(content=(ai or "")))
         context.append(HumanMessage(content=latest_message))
-        logger.debug("Chat context: %s", context)
+        log.debug("Chat context: %s", context)
 
         response = ""
         for chunk in llm.stream(context):
@@ -104,7 +100,7 @@ def inference(latest_message, history):
     # https://github.com/openai/openai-python/tree/e8e5a0dc7ccf2db19d7f81991ee0987f9c3ae375?tab=readme-ov-file#handling-errors
 
     except openai.BadRequestError as err:
-        logger.error("Received BadRequestError from backend API: %s", err)
+        log.error("Received BadRequestError from backend API: %s", err)
         message = err.response.json()["message"]
         if INCLUDE_SYSTEM_PROMPT:
             raise PossibleSystemPromptException()
@@ -115,12 +111,12 @@ def inference(latest_message, history):
 
     except openai.APIConnectionError as err:
         if not BACKEND_INITIALISED:
-            logger.info("Backend API not yet ready")
+            log.info("Backend API not yet ready")
             gr.Info(
                 "Backend not ready - model may still be initialising - please try again later."
             )
         else:
-            logger.error("Failed to connect to backend API: %s", err)
+            log.error("Failed to connect to backend API: %s", err)
             gr.Warning("Failed to connect to backend API.")
 
     except openai.InternalServerError as err:
@@ -130,7 +126,7 @@ def inference(latest_message, history):
 
     # Catch-all for unexpected exceptions
     except Exception as err:
-        logger.error("Unexpected error during inference: %s", err)
+        log.error("Unexpected error during inference: %s", err)
         raise gr.Error("Unexpected error encountered - see logs for details.")
 
 
@@ -150,7 +146,7 @@ def inference_wrapper(*args):
         for chunk in inference(*args):
             yield chunk
     except PossibleSystemPromptException:
-        logger.warning("Disabling system prompt and retrying previous request")
+        log.warning("Disabling system prompt and retrying previous request")
         INCLUDE_SYSTEM_PROMPT = False
         for chunk in inference(*args):
             yield chunk
@@ -179,7 +175,7 @@ def inference_wrapper(*args):
     css=settings.css_overrides,
     js=settings.custom_javascript,
 )
-logger.debug("Gradio chat interface config: %s", app.config)
+log.debug("Gradio chat interface config: %s", app.config)
 app.queue(
     default_concurrency_limit=10,
 ).launch(server_name=settings.host_address)

web-apps/chat/requirements.txt

@@ -4,4 +4,5 @@ openai
 langchain
 langchain_openai
 pydantic
+structlog
 ../utils

web-apps/image-analysis/app.py

@@ -1,20 +1,17 @@
 import base64
+import gradio as gr
 import logging
 import requests
-
-import gradio as gr
+import utils
 
 from typing import List, Dict
 from io import BytesIO
 from PIL import Image
 from pydantic import BaseModel, ConfigDict
 from urllib.parse import urljoin
 
-from utils import load_settings, LLMParams
 
-logging.basicConfig()
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.INFO)
+log = utils.get_logger()
 
 
 class PromptExample(BaseModel):
@@ -30,7 +27,7 @@ class AppSettings(BaseModel):
     page_title: str
     page_description: str
     examples: List[PromptExample]
-    llm_params: LLMParams | None
+    llm_params: utils.LLMParams | None
     # Theme customisation
     theme_params: Dict[str, str | list]
     theme_params_extended: Dict[str, str]
@@ -40,8 +37,8 @@ class AppSettings(BaseModel):
     model_config = ConfigDict(protected_namespaces=(), extra="forbid")
 
 
-settings = AppSettings(**load_settings())
-logger.info(settings)
+settings = AppSettings(**utils.load_settings())
+log.info(settings)
 
 
 # TODO: Rewrite this to stream output?
@@ -78,6 +75,7 @@ def analyze_image(image_url, prompt):
             payload["extra_body"] = {
                 "top_k": settings.llm_params.top_k,
             }
+        log.debug("Request payload: %s", payload)
 
         # Make the API call to the vision model
         headers = {"Content-Type": "application/json"}
@@ -86,7 +84,16 @@ def analyze_image(image_url, prompt):
             json=payload,
             headers=headers,
         )
-        response.raise_for_status()
+        log.debug("Request payload: %s", payload)
+        try:
+            response.raise_for_status()
+        except Exception as e:
+            log.debug(
+                "Received HTTP %s response with content: %s",
+                response.status_code,
+                response.json(),
+            )
+            raise e
 
         # Extract and return the model's response
         result = response.json()

web-apps/image-analysis/requirements.txt

@@ -3,4 +3,5 @@ requests
 gradio<5
 gradio_client
 pydantic
+structlog
 ../utils

web-apps/utils/utils.py

@@ -3,16 +3,29 @@
 #####
 
 import logging
+import os
 import pathlib
+import structlog
 import yaml
 from typing import Annotated
 from pydantic import BaseModel, ConfigDict, PositiveInt, Field
 
-logging.basicConfig()
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.INFO)
+LOG_LEVELS = {
+    "debug": logging.DEBUG,
+    "info": logging.INFO,
+    "warn": logging.WARN,
+    "error": logging.ERROR,
+}
 
 
+def get_logger():
+    # Allow overwriting log level via env var
+    log_level = LOG_LEVELS[os.environ.get("PYTHON_GRADIO_LOG_LEVEL", "info").lower()]
+    structlog.configure(wrapper_class=structlog.make_filtering_bound_logger(log_level))
+    return structlog.get_logger()
+
+log = get_logger()
+
 class LLMParams(BaseModel):
     """
     Parameters for vLLM API requests. For details see
@@ -37,7 +50,7 @@ def get_k8s_namespace():
     try:
         current_k8s_namespace = open(NAMESPACE_FILE_PATH).read()
         return current_k8s_namespace
-    except FileNotFoundError as err:
+    except FileNotFoundError:
         return None
 
 
@@ -46,7 +59,7 @@ def api_address_in_cluster():
     if k8s_ns:
         return f"http://llm-backend.{k8s_ns}.svc"
     else:
-        logger.warning(
+        log.warning(
             "Failed to determine k8s namespace from %s - assuming non-kubernetes environment.",
             NAMESPACE_FILE_PATH,
         )
@@ -77,7 +90,7 @@ def load_settings() -> dict:
     # Sanity checks on settings
     unused_overrides = [k for k in overrides.keys() if k not in defaults.keys()]
     if unused_overrides:
-        logger.warning(
+        log.warning(
             f"Overrides {unused_overrides} not part of default settings so may be ignored."
             "Please check for typos"
         )