mlrun 1.10.0rc16__py3-none-any.whl → 1.10.0rc42__py3-none-any.whl

mlrun/runtimes/base.py

@@ -142,9 +142,6 @@ class FunctionSpec(ModelObj):
     def build(self, build):
         self._build = self._verify_dict(build, "build", ImageBuilder)
 
-    def enrich_function_preemption_spec(self):
-        pass
-
     def validate_service_account(self, allowed_service_accounts):
         pass
 
@@ -379,7 +376,12 @@ class BaseRuntime(ModelObj):
                              This ensures latest code changes are executed. This argument must be used in
                              conjunction with the local=True argument.
         :param output_path:    Default artifact output path.
-        :param retry:          Retry configuration for the run, can be a dict or an instance of mlrun.model.Retry.
+        :param retry:          Retry configuration for the run, can be a dict or an instance of
+                               :py:class:`~mlrun.model.Retry`.
+                               The `count` field in the `Retry` object specifies the number of retry attempts.
+                               If `count=0`, the run will not be retried.
+                               The `backoff` field specifies the retry backoff strategy between retry attempts.
+                               If not provided, the default backoff delay is 30 seconds.
         :return: Run context object (RunObject) with run metadata, results and status
         """
         if artifact_path or out_path:
@@ -391,6 +393,7 @@ class BaseRuntime(ModelObj):
                 FutureWarning,
             )
         output_path = output_path or out_path or artifact_path
+
         launcher = mlrun.launcher.factory.LauncherFactory().create_launcher(
             self._is_remote, local=local, **launcher_kwargs
         )
@@ -446,15 +449,20 @@ class BaseRuntime(ModelObj):
         :param runobj: Run context object (RunObject) with run metadata and status
         :return: Dictionary with all the variables that could be parsed
         """
+        active_project = self.metadata.project or config.active_project
         runtime_env = {
-            "MLRUN_ACTIVE_PROJECT": self.metadata.project or config.active_project
+            mlrun_constants.MLRUN_ACTIVE_PROJECT: active_project,
+            # TODO: Remove this in 1.12.0 as MLRUN_DEFAULT_PROJECT is deprecated and should not be injected anymore
+            "MLRUN_DEFAULT_PROJECT": active_project,
         }
         if runobj:
             runtime_env["MLRUN_EXEC_CONFIG"] = runobj.to_json(
                 exclude_notifications_params=True
             )
             if runobj.metadata.project:
-                runtime_env["MLRUN_ACTIVE_PROJECT"] = runobj.metadata.project
+                runtime_env[mlrun_constants.MLRUN_ACTIVE_PROJECT] = (
+                    runobj.metadata.project
+                )
             if runobj.spec.verbose:
                 runtime_env["MLRUN_LOG_LEVEL"] = "DEBUG"
         if config.httpdb.api_url:

mlrun/runtimes/daskjob.py

@@ -541,6 +541,7 @@ class DaskCluster(KubejobRuntime):
             notifications=notifications,
             returns=returns,
             state_thresholds=state_thresholds,
+            retry=retry,
             **launcher_kwargs,
         )
 

mlrun/runtimes/local.py

@@ -29,12 +29,12 @@ from os import environ, remove
 from pathlib import Path
 from subprocess import PIPE, Popen
 from sys import executable
+from typing import Optional
 
 from nuclio import Event
 
 import mlrun
 import mlrun.common.constants as mlrun_constants
-import mlrun.common.runtimes.constants
 from mlrun.lists import RunList
 
 from ..errors import err_to_str
@@ -201,9 +201,12 @@ class LocalRuntime(BaseRuntime, ParallelRunner):
     kind = "local"
     _is_remote = False
 
-    def to_job(self, image=""):
+    def to_job(self, image="", func_name: Optional[str] = None):
         struct = self.to_dict()
         obj = KubejobRuntime.from_dict(struct)
+        obj.kind = "job"  # Ensure kind is set to 'job' for KubejobRuntime
+        if func_name:
+            obj.metadata.name = func_name
         if image:
             obj.spec.image = image
         return obj

mlrun/runtimes/mounts.py

@@ -14,8 +14,11 @@
 
 import os
 import typing
+import warnings
 from collections import namedtuple
 
+import mlrun.common.secrets
+import mlrun.errors
 from mlrun.config import config
 from mlrun.config import config as mlconf
 from mlrun.errors import MLRunInvalidArgumentError
@@ -247,10 +250,22 @@ def mount_s3(
     def _use_s3_cred(runtime: "KubeResource"):
         _access_key = aws_access_key or os.environ.get(prefix + "AWS_ACCESS_KEY_ID")
         _secret_key = aws_secret_key or os.environ.get(prefix + "AWS_SECRET_ACCESS_KEY")
-        _endpoint_url = endpoint_url or os.environ.get(prefix + "S3_ENDPOINT_URL")
+
+        # Check for endpoint URL with backward compatibility
+        _endpoint_url = endpoint_url or os.environ.get(prefix + "AWS_ENDPOINT_URL_S3")
+        if not _endpoint_url:
+            # Check for deprecated environment variable
+            _endpoint_url = os.environ.get(prefix + "S3_ENDPOINT_URL")
+            if _endpoint_url:
+                warnings.warn(
+                    "S3_ENDPOINT_URL is deprecated in 1.10.0 and will be removed in 1.12.0, "
+                    "use AWS_ENDPOINT_URL_S3 instead.",
+                    # TODO: Remove this in 1.12.0
+                    FutureWarning,
+                )
 
         if _endpoint_url:
-            runtime.set_env(prefix + "S3_ENDPOINT_URL", _endpoint_url)
+            runtime.set_env(prefix + "AWS_ENDPOINT_URL_S3", _endpoint_url)
         if aws_region:
             runtime.set_env(prefix + "AWS_REGION", aws_region)
         if non_anonymous:
@@ -399,6 +414,9 @@ def mount_secret(
                          the specified paths, and unlisted keys will not be
                          present."""
 
+    if secret_name:
+        mlrun.common.secrets.validate_not_forbidden_secret(secret_name.strip())
+
     def _mount_secret(runtime: "KubeResource"):
         # Define the secret volume source
         secret_volume_source = {

mlrun/runtimes/nuclio/__init__.py

@@ -16,6 +16,7 @@ from .serving import ServingRuntime, new_v2_model_server  # noqa
 from .nuclio import nuclio_init_hook  # noqa
 from .function import (
     min_nuclio_versions,
+    multiple_port_sidecar_is_supported,
     RemoteRuntime,
 )  # noqa
 from .api_gateway import APIGateway

mlrun/runtimes/nuclio/application/application.py

@@ -22,19 +22,23 @@ import mlrun.errors
 import mlrun.run
 from mlrun.common.runtimes.constants import NuclioIngressAddTemplatedIngressModes
 from mlrun.runtimes import RemoteRuntime
-from mlrun.runtimes.nuclio import min_nuclio_versions
+from mlrun.runtimes.nuclio import (
+    min_nuclio_versions,
+    multiple_port_sidecar_is_supported,
+)
 from mlrun.runtimes.nuclio.api_gateway import (
     APIGateway,
     APIGatewayMetadata,
     APIGatewaySpec,
 )
 from mlrun.runtimes.nuclio.function import NuclioSpec, NuclioStatus
-from mlrun.utils import logger, update_in
+from mlrun.utils import is_valid_port, logger, update_in
 
 
 class ApplicationSpec(NuclioSpec):
     _dict_fields = NuclioSpec._dict_fields + [
         "internal_application_port",
+        "application_ports",
     ]
 
     def __init__(
@@ -78,7 +82,12 @@ class ApplicationSpec(NuclioSpec):
         add_templated_ingress_host_mode=None,
         state_thresholds=None,
         disable_default_http_trigger=None,
+        serving_spec=None,
+        graph=None,
+        parameters=None,
+        track_models=None,
         internal_application_port=None,
+        application_ports=None,
     ):
         super().__init__(
             command=command,
@@ -118,6 +127,10 @@ class ApplicationSpec(NuclioSpec):
             security_context=security_context,
             service_type=service_type,
             add_templated_ingress_host_mode=add_templated_ingress_host_mode,
+            serving_spec=serving_spec,
+            graph=graph,
+            parameters=parameters,
+            track_models=track_models,
             state_thresholds=state_thresholds,
             disable_default_http_trigger=disable_default_http_trigger,
         )
@@ -126,11 +139,60 @@ class ApplicationSpec(NuclioSpec):
         self.min_replicas = min_replicas or 1
         self.max_replicas = max_replicas or 1
 
+        # initializing internal application port and application ports
+        self._internal_application_port = None
+        self._application_ports = []
+
+        application_ports = application_ports or []
+
+        # if internal_application_port is not provided, use the first application port
+        if not internal_application_port and len(application_ports) > 0:
+            internal_application_port = application_ports[0]
+
+        # the port of application sidecar to which traffic will be routed from a nuclio function
         self.internal_application_port = (
             internal_application_port
             or mlrun.mlconf.function.application.default_sidecar_internal_port
         )
 
+        # all exposed ports by the application sidecar
+        self.application_ports = application_ports
+
+    @property
+    def application_ports(self):
+        return self._application_ports
+
+    @application_ports.setter
+    def application_ports(self, ports):
+        """
+        Set the application ports for the application sidecar.
+        The internal application port is always included and always first.
+        """
+        # Handle None / single int
+        if ports is None:
+            ports = []
+        elif isinstance(ports, int):
+            ports = [ports]
+        elif not isinstance(ports, list):
+            raise mlrun.errors.MLRunInvalidArgumentError(
+                "Application ports must be a list of integers"
+            )
+
+        # Validate and normalize
+        cleaned_ports = []
+        for port in ports:
+            is_valid_port(port, raise_on_error=True)
+            if port != self.internal_application_port:
+                cleaned_ports.append(port)
+
+        application_ports = [self.internal_application_port] + cleaned_ports
+
+        # ensure multiple ports are supported in Nuclio
+        if len(application_ports) > 1:
+            multiple_port_sidecar_is_supported()
+
+        self._application_ports = application_ports
+
     @property
     def internal_application_port(self):
         return self._internal_application_port
@@ -138,10 +200,20 @@ class ApplicationSpec(NuclioSpec):
     @internal_application_port.setter
     def internal_application_port(self, port):
         port = int(port)
-        if port < 0 or port > 65535:
-            raise ValueError("Port must be in the range 0-65535")
+        is_valid_port(port, raise_on_error=True)
         self._internal_application_port = port
 
+        # If when internal application port is being set, length of self._application_ports is 1,
+        # it means that it consist of [old_port] only
+        # so in this case, we rewrite the list completely, by setting value to [new_value]
+        if len(self.application_ports) == 1:
+            self._application_ports = [port]
+            return
+
+        # when setting new internal application port, ensure that it is included in the application ports
+        # it just triggers setter logic, so setting to the same value is a no-op
+        self.application_ports = self._application_ports
+
 
 class ApplicationStatus(NuclioStatus):
     def __init__(
@@ -222,6 +294,32 @@ class ApplicationRuntime(RemoteRuntime):
     def set_internal_application_port(self, port: int):
         self.spec.internal_application_port = port
 
+    def with_sidecar(
+        self,
+        name: typing.Optional[str] = None,
+        image: typing.Optional[str] = None,
+        ports: typing.Optional[typing.Union[int, list[int]]] = None,
+        command: typing.Optional[str] = None,
+        args: typing.Optional[list[str]] = None,
+    ):
+        # wraps with_sidecar just to set the application ports
+        super().with_sidecar(
+            name=name,
+            image=image,
+            ports=ports,
+            command=command,
+            args=args,
+        )
+
+        if ports:
+            if self.spec.internal_application_port != ports[0]:
+                logger.info(
+                    f"Setting internal application port to the first port from the sidecar: {ports[0]}. "
+                    f"If this is not intended, please set the internal_application_port explicitly."
+                )
+                self.spec.internal_application_port = ports[0]
+            self.spec.application_ports = ports
+
     def pre_deploy_validation(self):
         super().pre_deploy_validation()
         if not self.spec.config.get("spec.sidecars"):
@@ -302,6 +400,7 @@ class ApplicationRuntime(RemoteRuntime):
 
         :return: The default API gateway URL if created or True if the function is ready (deployed)
         """
+
         if (self.requires_build() and not self.spec.image) or force_build:
             self._fill_credentials()
             self._build_application_image(
@@ -315,8 +414,7 @@ class ApplicationRuntime(RemoteRuntime):
                 show_on_failure=show_on_failure,
             )
 
-        # This is a class method that accepts a function instance, so we pass self as the function instance
-        self._ensure_reverse_proxy_configurations(self)
+        self._ensure_reverse_proxy_configurations()
         self._configure_application_sidecar()
 
         # We only allow accessing the application via the API Gateway
@@ -431,6 +529,7 @@ class ApplicationRuntime(RemoteRuntime):
         ssl_redirect: typing.Optional[bool] = None,
         set_as_default: bool = False,
         gateway_timeout: typing.Optional[int] = None,
+        port: typing.Optional[int] = None,
     ):
         """
         Create the application API gateway. Once the application is deployed, the API gateway can be created.
@@ -447,6 +546,8 @@ class ApplicationRuntime(RemoteRuntime):
         :param set_as_default:          Set the API gateway as the default for the application (`status.api_gateway`)
         :param gateway_timeout:         nginx ingress timeout in sec (request timeout, when will the gateway return an
                                         error)
+        :param port:                    The API gateway port, used only when direct_port_access=True
+
         :return:                        The API gateway URL
         """
         if not name:
@@ -467,7 +568,15 @@ class ApplicationRuntime(RemoteRuntime):
                 "Authentication credentials not provided"
             )
 
-        ports = self.spec.internal_application_port if direct_port_access else []
+        if direct_port_access and port:
+            logger.warning(
+                "Ignoring 'port' because 'direct_port_access' is enabled. "
+                "The 'port' setting is only applicable when 'direct_port_access' is disabled."
+            )
+
+        ports = (
+            port or self.spec.internal_application_port if direct_port_access else []
+        )
 
         api_gateway = APIGateway(
             APIGatewayMetadata(
@@ -595,6 +704,12 @@ class ApplicationRuntime(RemoteRuntime):
         """
         # create a function that includes only the reverse proxy, without the application
 
+        if not mlrun.get_current_project(silent=True):
+            raise mlrun.errors.MLRunMissingProjectError(
+                "An active project is required to run deploy_reverse_proxy_image(). "
+                "Use `mlrun.get_or_create_project()` or set an active project first."
+            )
+
         reverse_proxy_func = mlrun.run.new_function(
             name="reverse-proxy-temp", kind="remote"
         )
@@ -684,27 +799,42 @@ class ApplicationRuntime(RemoteRuntime):
             with_mlrun=with_mlrun,
         )
 
-    @staticmethod
-    def _ensure_reverse_proxy_configurations(function: RemoteRuntime):
-        if function.spec.build.functionSourceCode or function.status.container_image:
+    def _ensure_reverse_proxy_configurations(self):
+        # If an HTTP trigger already exists in the spec,
+        # it means the user explicitly defined a custom configuration,
+        # so, skip automatic creation.
+        skip_http_trigger_creation = False
+        for key, value in self.spec.config.items():
+            if key.startswith("spec.triggers"):
+                if isinstance(value, dict):
+                    if value.get("kind") == "http":
+                        skip_http_trigger_creation = True
+                        break
+        if not skip_http_trigger_creation:
+            self.with_http(
+                workers=mlrun.mlconf.function.application.default_worker_number,
+                trigger_name="application-http",
+            )
+
+        if self.spec.build.functionSourceCode or self.status.container_image:
             return
 
         filename, handler = ApplicationRuntime.get_filename_and_handler()
         name, spec, code = nuclio.build_file(
             filename,
-            name=function.metadata.name,
+            name=self.metadata.name,
             handler=handler,
         )
-        function.spec.function_handler = mlrun.utils.get_in(spec, "spec.handler")
-        function.spec.build.functionSourceCode = mlrun.utils.get_in(
+        self.spec.function_handler = mlrun.utils.get_in(spec, "spec.handler")
+        self.spec.build.functionSourceCode = mlrun.utils.get_in(
             spec, "spec.build.functionSourceCode"
         )
-        function.spec.nuclio_runtime = mlrun.utils.get_in(spec, "spec.runtime")
+        self.spec.nuclio_runtime = mlrun.utils.get_in(spec, "spec.runtime")
 
         # default the reverse proxy logger level to info
         logger_sinks_key = "spec.loggerSinks"
-        if not function.spec.config.get(logger_sinks_key):
-            function.set_config(
+        if not self.spec.config.get(logger_sinks_key):
+            self.set_config(
                 logger_sinks_key, [{"level": "info", "sink": "myStdoutLoggerSink"}]
             )
 
@@ -728,7 +858,7 @@ class ApplicationRuntime(RemoteRuntime):
         self.with_sidecar(
             name=self.status.sidecar_name,
             image=self.status.application_image,
-            ports=self.spec.internal_application_port,
+            ports=self.spec.application_ports,
             command=self.spec.command,
             args=self.spec.args,
         )

mlrun/runtimes/nuclio/function.py

@@ -16,6 +16,7 @@ import asyncio
 import copy
 import json
 import typing
+import warnings
 from datetime import datetime
 from time import sleep
 
@@ -29,6 +30,7 @@ from kubernetes import client
 from nuclio.deploy import find_dashboard_url, get_deploy_status
 from nuclio.triggers import V3IOStreamTrigger
 
+import mlrun.common.constants
 import mlrun.db
 import mlrun.errors
 import mlrun.k8s_utils
@@ -422,6 +424,18 @@ class RemoteRuntime(KubeResource):
                 )
         """
         self.spec.build.source = source
+
+        code = (
+            self.spec.build.functionSourceCode if hasattr(self.spec, "build") else None
+        )
+        if code:
+            # Warn and clear any inline code so the archive is actually used
+            logger.warning(
+                "Cannot specify both code and source archive. Removing the code so the provided "
+                "source archive will be used instead."
+            )
+            self.spec.build.functionSourceCode = None
+
         # update handler in function_handler if needed
         if handler:
             self.spec.function_handler = handler
@@ -830,7 +844,8 @@ class RemoteRuntime(KubeResource):
     def _get_runtime_env(self):
         # for runtime specific env var enrichment (before deploy)
         runtime_env = {
-            "MLRUN_ACTIVE_PROJECT": self.metadata.project or mlconf.active_project,
+            mlrun.common.constants.MLRUN_ACTIVE_PROJECT: self.metadata.project
+            or mlconf.active_project,
         }
         if mlconf.httpdb.api_url:
             runtime_env["MLRUN_DBPATH"] = mlconf.httpdb.api_url
@@ -966,24 +981,6 @@ class RemoteRuntime(KubeResource):
         self._mock_server = None
 
         if "://" not in path:
-            if not self.status.address:
-                # here we check that if default http trigger is disabled, function contains a custom http trigger
-                # Otherwise, the function is not invokable, so we raise an error
-                if (
-                    not self._trigger_of_kind_exists(kind="http")
-                    and self.spec.disable_default_http_trigger
-                ):
-                    raise mlrun.errors.MLRunPreconditionFailedError(
-                        "Default http trigger creation is disabled and there is no any other custom http trigger, "
-                        "so function can not be invoked via http. Either enable default http trigger creation or "
-                        "create custom http trigger"
-                    )
-                state, _, _ = self._get_state()
-                if state not in ["ready", "scaledToZero"]:
-                    logger.warning(f"Function is in the {state} state")
-                if not self.status.address:
-                    raise ValueError("no function address first run .deploy()")
-
             path = self._resolve_invocation_url(path, force_external_address)
 
         if headers is None:
@@ -1043,6 +1040,9 @@ class RemoteRuntime(KubeResource):
             sidecar["image"] = image
 
         ports = mlrun.utils.helpers.as_list(ports)
+        if len(ports) > 1:
+            mlrun.runtimes.nuclio.multiple_port_sidecar_is_supported()
+
         # according to RFC-6335, port name should be less than 15 characters,
         # so we truncate it if needed and leave room for the index
         port_name = name[:13].rstrip("-_") if len(name) > 13 else name
@@ -1223,19 +1223,54 @@ class RemoteRuntime(KubeResource):
         # internal / external invocation urls is a nuclio >= 1.6.x feature
         # try to infer the invocation url from the internal and if not exists, use external.
         # $$$$ we do not want to use the external invocation url (e.g.: ingress, nodePort, etc.)
+
+        # if none of urls is set, function was deployed with watch=False
+        # and status wasn't fetched with Nuclio
+        # _get_state fetches the state and updates url
+        if (
+            not self.status.address
+            and not self.status.internal_invocation_urls
+            and not self.status.external_invocation_urls
+        ):
+            state, _, _ = self._get_state()
+            if state not in ["ready", "scaledToZero"]:
+                logger.warning(f"Function is in the {state} state")
+
+        # prefer internal invocation url if running inside k8s cluster
         if (
             not force_external_address
             and self.status.internal_invocation_urls
             and mlrun.k8s_utils.is_running_inside_kubernetes_cluster()
         ):
-            return mlrun.utils.helpers.join_urls(
+            url = mlrun.utils.helpers.join_urls(
                 f"http://{self.status.internal_invocation_urls[0]}", path
             )
+            logger.debug(
+                f"Using internal invocation url {url}. Make sure you have network access to the k8s cluster. "
+                f"Otherwise, set force_external_address to True"
+            )
+            return url
 
         if self.status.external_invocation_urls:
             return mlrun.utils.helpers.join_urls(
                 f"http://{self.status.external_invocation_urls[0]}", path
             )
+
+        if not self.status.address:
+            # if there is no address
+            # here we check that if default http trigger is disabled, function contains a custom http trigger
+            # Otherwise, the function is not invokable, so we raise an error
+            if (
+                not self._trigger_of_kind_exists(kind="http")
+                and self.spec.disable_default_http_trigger
+            ):
+                raise mlrun.errors.MLRunPreconditionFailedError(
+                    "Default http trigger creation is disabled and there is no any other custom http trigger, "
+                    "so function can not be invoked via http. Either enable default http trigger creation or "
+                    "create custom http trigger"
+                )
+            else:
+                raise ValueError("no function address first run .deploy()")
         else:
             return mlrun.utils.helpers.join_urls(f"http://{self.status.address}", path)
 
@@ -1289,6 +1324,8 @@ class RemoteRuntime(KubeResource):
     def get_url(
         self,
         force_external_address: bool = False,
+        # leaving auth_info for BC
+        # TODO: remove in 1.12.0
         auth_info: AuthInfo = None,
     ):
         """
@@ -1299,13 +1336,12 @@ class RemoteRuntime(KubeResource):
 
         :return: returns function's url
         """
-        if not self.status.address:
-            state, _, _ = self._get_state(auth_info=auth_info)
-            if state != "ready" or not self.status.address:
-                raise ValueError(
-                    "no function address or not ready, first run .deploy()"
-                )
-
+        if auth_info:
+            warnings.warn(
+                "'auth_info' is deprecated in 1.10.0 and will be removed in 1.12.0.",
+                # TODO: Remove this in 1.12.0
+                FutureWarning,
+            )
         return self._resolve_invocation_url("", force_external_address)
 
     @staticmethod
@@ -1456,3 +1492,10 @@ def enrich_nuclio_function_from_headers(
         else []
     )
     func.status.container_image = headers.get("x-mlrun-container-image", "")
+
+
+@min_nuclio_versions("1.14.14")
+def multiple_port_sidecar_is_supported():
+    # multiple ports are supported from nuclio version 1.14.14
+    # this method exists only for running the min_nuclio_versions decorator
+    return True