OPIK-814 [SDK] Implement opik healthcheck CLI command / python function to validate the current...

sdks/python/src/opik/api_objects/opik_client.py

@@ -50,6 +50,7 @@ def __init__(
         host: Optional[str] = None,
         api_key: Optional[str] = None,
         _use_batching: bool = False,
+        _show_misconfiguration_message: bool = True,
     ) -> None:
         """
         Initialize an Opik object that can be used to log traces and spans manually to Opik server.
@@ -61,16 +62,22 @@ def __init__(
             api_key: The API key for Opik. This parameter is ignored for local installations.
             _use_batching: intended for internal usage in specific conditions only.
                 Enabling it is unsafe and can lead to data loss.
+            _show_misconfiguration_message: intended for internal usage in specific conditions only.
+                Print a warning message if the Opik server is not configured properly.
         Returns:
             None
         """
+
         config_ = config.get_from_user_inputs(
             project_name=project_name,
             workspace=workspace,
             url_override=host,
             api_key=api_key,
         )
-        config.is_misconfigured(config_, show_misconfiguration_message=True)
+
+        config_.is_misconfigured(
+            show_misconfiguration_message=_show_misconfiguration_message,
+        )
         self._config = config_
 
         self._workspace: str = config_.workspace
@@ -895,6 +902,7 @@ def create_prompt(
         Parameters:
             name: The name of the prompt.
             prompt: The template content of the prompt.
+            metadata: Optional metadata to be included in the prompt.
 
         Returns:
             A Prompt object containing details of the created or retrieved prompt.

sdks/python/src/opik/cli.py

@@ -5,8 +5,12 @@
 from importlib import metadata
 
 import click
-from opik.configurator import configure as opik_configure
-from opik.configurator import interactive_helpers
+from rich.console import Console
+
+from opik import healthcheck as opik_healthcheck
+from opik.configurator import configure as opik_configure, interactive_helpers
+
+console = Console()
 
 LOGGER = logging.getLogger(__name__)
 
@@ -169,5 +173,21 @@ def proxy(
     )  # Reduce uvicorn logging to keep output clean
 
 
+@cli.command(context_settings={"ignore_unknown_options": True})
+@click.option(
+    "--show-installed-packages",
+    is_flag=True,
+    default=False,
+    help="Print the list of installed packages to the console.",
+)
+def healthcheck(show_installed_packages: bool = True) -> None:
+    """
+    Performs a health check of the application, including validation of configuration,
+    verification of library installations, and checking the availability of the backend workspace.
+    Prints all relevant information to assist in debugging and diagnostics.
+    """
+    opik_healthcheck.run(show_installed_packages)
+
+
 if __name__ == "__main__":
     cli()

sdks/python/src/opik/config.py

@@ -238,72 +238,152 @@ def _set_url_override_from_api_key(self) -> "OpikConfig":
 
         return self
 
+    @property
+    def is_config_file_exists(self) -> bool:
+        """
+        Determines whether the configuration file exists at the specified path.
+        """
+        return self.config_file_fullpath.exists()
 
-def update_session_config(key: str, value: Any) -> None:
-    _SESSION_CACHE_DICT[key] = value
+    @property
+    def is_cloud_installation(self) -> bool:
+        """
+        Determine if the installation type is a cloud installation.
+        """
+        return url_helpers.get_base_url(self.url_override) == url_helpers.get_base_url(
+            OPIK_URL_CLOUD
+        )
 
+    def get_current_config_with_api_key_hidden(self) -> Dict[str, Any]:
+        """
+        Retrieves the current configuration with the API key value masked.
+        """
+        current_values = self.model_dump()
+        if current_values.get("api_key") is not None:
+            current_values["api_key"] = "*** HIDDEN ***"
+        return current_values
 
-def get_from_user_inputs(**user_inputs: Any) -> OpikConfig:
-    """
-    Instantiates an OpikConfig using provided user inputs.
-    """
-    cleaned_user_inputs = dict_utils.remove_none_from_dict(user_inputs)
+    def is_misconfigured(self, show_misconfiguration_message: bool = False) -> bool:
+        """
+        Determines if Opik configuration is misconfigured and optionally displays
+        a corresponding error message.
 
-    return OpikConfig(**cleaned_user_inputs)
+        Parameters:
+        show_misconfiguration_message : A flag indicating whether to display detailed error messages if the configuration
+            is determined to be misconfigured. Defaults to False.
+        """
 
+        is_misconfigured_flag, error_message = (
+            self.get_misconfiguration_validation_results()
+        )
 
-def is_misconfigured(
-    config: OpikConfig,
-    show_misconfiguration_message: bool = False,
-) -> bool:
-    """
-    Determines if the provided Opik configuration is misconfigured and optionally displays
-    a corresponding error message.
-
-    Parameters:
-    config: The configuration object containing settings such as URL overrides, workspace, API key,
-        and tracking options to be validated for misconfiguration.
-    show_misconfiguration_message : A flag indicating whether to display detailed error messages if the configuration
-        is determined to be misconfigured. Defaults to False.
-    """
+        if is_misconfigured_flag:
+            if show_misconfiguration_message:
+                print()
+                LOGGER.error(
+                    "========================\n"
+                    f"{error_message}\n"
+                    "==============================\n"
+                )
+            return True
 
-    cloud_installation = url_helpers.get_base_url(
-        config.url_override
-    ) == url_helpers.get_base_url(OPIK_URL_CLOUD)
-    localhost_installation = (
-        "localhost" in config.url_override
-    )  # does not detect all OSS installations
-    workspace_is_default = config.workspace == OPIK_WORKSPACE_DEFAULT_NAME
-    api_key_configured = config.api_key is not None
-    tracking_disabled = config.track_disable
-
-    if (
-        cloud_installation
-        and (not api_key_configured or workspace_is_default)
-        and not tracking_disabled
-    ):
-        if show_misconfiguration_message:
-            print()
-            LOGGER.error(
-                "========================\n"
+        return False
+
+    def is_misconfigured_for_cloud(self) -> Tuple[bool, Optional[str]]:
+        """
+        Determines if the current Opik configuration is misconfigured for cloud logging.
+
+        Returns:
+            Tuple[bool, Optional[str]]: A tuple where the first element is a boolean indicating if
+            the configuration is misconfigured for cloud logging, and the second element is either
+            an error message indicating the reason for misconfiguration or None.
+        """
+        workspace_is_default = self.workspace == OPIK_WORKSPACE_DEFAULT_NAME
+        api_key_configured = self.api_key is not None
+        tracking_disabled = self.track_disable
+
+        if (
+            self.is_cloud_installation
+            and (not api_key_configured or workspace_is_default)
+            and not tracking_disabled
+        ):
+            error_message = (
                 "The workspace and API key must be specified to log data to https://www.comet.com/opik.\n"
                 "You can use `opik configure` CLI command to configure your environment for logging.\n"
                 "See the configuration details in the docs: https://www.comet.com/docs/opik/tracing/sdk_configuration.\n"
-                "==============================\n"
             )
-        return True
+            return True, error_message
+
+        return False, None
+
+    def is_misconfigured_for_local(self) -> Tuple[bool, Optional[str]]:
+        """
+        Determines if the current setup is misconfigured for a local open-source installation.
 
-    if localhost_installation and not workspace_is_default and not tracking_disabled:
-        if show_misconfiguration_message:
-            print()
-            LOGGER.error(
-                "========================\n"
+        Returns:
+            Tuple[bool, Optional[str]]: A tuple where the first element is a boolean indicating if
+            the configuration is misconfigured for local logging, and the second element is either
+            an error message indicating the reason for misconfiguration or None.
+        """
+        localhost_installation = (
+            "localhost" in self.url_override
+        )  # does not detect all OSS installations
+        workspace_is_default = self.workspace == OPIK_WORKSPACE_DEFAULT_NAME
+        tracking_disabled = self.track_disable
+
+        if (
+            localhost_installation
+            and not workspace_is_default
+            and not tracking_disabled
+        ):
+            error_message = (
                 "Open source installations do not support workspace specification. Only `default` is available.\n"
                 "See the configuration details in the docs: https://www.comet.com/docs/opik/tracing/sdk_configuration\n"
                 "If you need advanced workspace management - you may consider using our cloud offer (https://www.comet.com/site/pricing/)\n"
                 "or contact our team for purchasing and setting up a self-hosted installation.\n"
-                "==============================\n"
             )
-        return True
+            return True, error_message
+
+        return False, None
+
+    def get_misconfiguration_validation_results(self) -> Tuple[bool, Optional[str]]:
+        """
+        Validates the current configuration and identifies any misconfigurations
+        for either cloud or local environments. This method checks both cloud
+        and local configurations and determines the validity of each, returning
+        a boolean indicator of success or failure and an optional error message
+        if there is an issue.
+
+        Returns:
+            Tuple[bool, Optional[str]]: A tuple where the first element indicates
+            whether the configuration is misconfigured (True for misconfigured, False for valid).
+            The second element is an optional string that contains
+            an error message if there is a configuration issue, or None if the
+            configuration is valid.
+        """
+        is_misconfigured_for_cloud_flag, error_message = (
+            self.is_misconfigured_for_cloud()
+        )
+        if is_misconfigured_for_cloud_flag:
+            return True, error_message
+
+        is_misconfigured_for_local_flag, error_message = (
+            self.is_misconfigured_for_local()
+        )
+        if is_misconfigured_for_local_flag:
+            return True, error_message
+
+        return False, None
 
-    return False
+
+def update_session_config(key: str, value: Any) -> None:
+    _SESSION_CACHE_DICT[key] = value
+
+
+def get_from_user_inputs(**user_inputs: Any) -> OpikConfig:
+    """
+    Instantiates an OpikConfig using provided user inputs.
+    """
+    cleaned_user_inputs = dict_utils.remove_none_from_dict(user_inputs)
+
+    return OpikConfig(**cleaned_user_inputs)

sdks/python/src/opik/environment.py

@@ -1,12 +1,12 @@
+import functools
 import getpass
 import logging
 import os
 import platform
 import socket
 import sys
-import functools
-
-from typing import Literal
+from importlib import metadata
+from typing import Dict, Literal
 
 import opik.config
 from opik import url_helpers
@@ -129,3 +129,14 @@ def in_colab() -> bool:
 
     ipy = IPython.get_ipython()
     return "google.colab" in str(ipy)
+
+
+@functools.lru_cache
+def get_installed_packages() -> Dict[str, str]:
+    """
+    Retrieve a dictionary of installed packages with their versions.
+    """
+    installed_packages = {
+        pkg.metadata["Name"]: pkg.version for pkg in metadata.distributions()
+    }
+    return installed_packages

sdks/python/src/opik/evaluation/metrics/base_metric.py

@@ -39,7 +39,7 @@ def __init__(self, name: str, track: bool = True) -> None:
 
         config = opik_config.OpikConfig()
 
-        if track and opik_config.is_misconfigured(config) is False:
+        if track and config.is_misconfigured() is False:
             self.score = opik.track(name=self.name)(self.score)  # type: ignore
             self.ascore = opik.track(name=self.name)(self.ascore)  # type: ignore
 

sdks/python/src/opik/evaluation/models/litellm/opik_monitor.py

@@ -35,7 +35,7 @@ def enabled_in_config() -> bool:
 @functools.lru_cache
 def opik_is_misconfigured() -> bool:
     config_ = config.OpikConfig()
-    return config.is_misconfigured(config_)
+    return config_.is_misconfigured()
 
 
 def _add_span_metadata_to_params(params: Dict[str, Any]) -> Dict[str, Any]:

sdks/python/src/opik/forwarding_server/app.py

@@ -178,25 +178,25 @@ async def chat_completions(request: Request) -> Any:
                     )
                     response = await client.chat.completions.create(**body)
                     return JSONResponse(response.model_dump())
-            except APIConnectionError as e:
+            except APITimeoutError as e:
                 forward_logger.error(
-                    f"[dim]{request.state.request_id}[/] Failed to connect to LLM server: {str(e)}"
+                    f"[dim]{request.state.request_id}[/] LLM server timeout: {str(e)}"
                 )
                 raise HTTPException(
-                    status_code=503,
+                    status_code=504,
                     detail={
-                        "error": f"Failed to connect to LLM server. Is it running at {llm_server_host}?",
+                        "error": "LLM server took too long to respond",
                         "request_id": request.state.request_id,
                     },
                 )
-            except APITimeoutError as e:
+            except APIConnectionError as e:
                 forward_logger.error(
-                    f"[dim]{request.state.request_id}[/] LLM server timeout: {str(e)}"
+                    f"[dim]{request.state.request_id}[/] Failed to connect to LLM server: {str(e)}"
                 )
                 raise HTTPException(
-                    status_code=504,
+                    status_code=503,
                     detail={
-                        "error": "LLM server took too long to respond",
+                        "error": f"Failed to connect to LLM server. Is it running at {llm_server_host}?",
                         "request_id": request.state.request_id,
                     },
                 )