#
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
# Note: Any AirflowException raised is expected to cause the TaskInstance
# to be marked in an ERROR state
"""Exceptions used by Airflow."""
from __future__ import annotations
import warnings
from datetime import timedelta
from http import HTTPStatus
from typing import TYPE_CHECKING, Any, NamedTuple
from airflow.utils.trigger_rule import TriggerRule
if TYPE_CHECKING:
from collections.abc import Sized
from airflow.models import DagRun
[docs]class AirflowException(Exception):
"""
Base class for all Airflow's errors.
Each custom exception should be derived from this class.
"""
[docs] status_code = HTTPStatus.INTERNAL_SERVER_ERROR
[docs] def serialize(self):
cls = self.__class__
return f"{cls.__module__}.{cls.__name__}", (str(self),), {}
[docs]class AirflowBadRequest(AirflowException):
"""Raise when the application or server cannot handle the request."""
[docs] status_code = HTTPStatus.BAD_REQUEST
[docs]class AirflowNotFoundException(AirflowException):
"""Raise when the requested object/resource is not available in the system."""
[docs] status_code = HTTPStatus.NOT_FOUND
[docs]class AirflowConfigException(AirflowException):
"""Raise when there is configuration problem."""
[docs]class AirflowSensorTimeout(AirflowException):
"""Raise when there is a timeout on sensor polling."""
[docs]class AirflowRescheduleException(AirflowException):
"""
Raise when the task should be re-scheduled at a later time.
:param reschedule_date: The date when the task should be rescheduled
"""
def __init__(self, reschedule_date):
super().__init__()
self.reschedule_date = reschedule_date
[docs] def serialize(self):
cls = self.__class__
return f"{cls.__module__}.{cls.__name__}", (), {"reschedule_date": self.reschedule_date}
[docs]class InvalidStatsNameException(AirflowException):
"""Raise when name of the stats is invalid."""
# Important to inherit BaseException instead of AirflowException->Exception, since this Exception is used
# to explicitly interrupt ongoing task. Code that does normal error-handling should not treat
# such interrupt as an error that can be handled normally. (Compare with KeyboardInterrupt)
[docs]class AirflowTaskTimeout(BaseException):
"""Raise when the task execution times-out."""
[docs]class AirflowTaskTerminated(BaseException):
"""Raise when the task execution is terminated."""
[docs]class AirflowWebServerTimeout(AirflowException):
"""Raise when the web server times out."""
[docs]class AirflowSkipException(AirflowException):
"""Raise when the task should be skipped."""
[docs]class AirflowFailException(AirflowException):
"""Raise when the task should be failed without retrying."""
[docs]class AirflowOptionalProviderFeatureException(AirflowException):
"""Raise by providers when imports are missing for optional provider features."""
class AirflowInternalRuntimeError(BaseException):
"""
Airflow Internal runtime error.
Indicates that something really terrible happens during the Airflow execution.
:meta private:
"""
[docs]class XComNotFound(AirflowException):
"""Raise when an XCom reference is being resolved against a non-existent XCom."""
def __init__(self, dag_id: str, task_id: str, key: str) -> None:
super().__init__()
self.dag_id = dag_id
self.task_id = task_id
self.key = key
[docs] def __str__(self) -> str:
return f'XComArg result from {self.task_id} at {self.dag_id} with key="{self.key}" is not found!'
[docs] def serialize(self):
cls = self.__class__
return (
f"{cls.__module__}.{cls.__name__}",
(),
{"dag_id": self.dag_id, "task_id": self.task_id, "key": self.key},
)
[docs]class UnmappableOperator(AirflowException):
"""Raise when an operator is not implemented to be mappable."""
[docs]class UnmappableXComTypePushed(AirflowException):
"""Raise when an unmappable type is pushed as a mapped downstream's dependency."""
def __init__(self, value: Any, *values: Any) -> None:
super().__init__(value, *values)
[docs] def __str__(self) -> str:
typename = type(self.args[0]).__qualname__
for arg in self.args[1:]:
typename = f"{typename}[{type(arg).__qualname__}]"
return f"unmappable return type {typename!r}"
[docs]class UnmappableXComLengthPushed(AirflowException):
"""Raise when the pushed value is too large to map as a downstream's dependency."""
def __init__(self, value: Sized, max_length: int) -> None:
super().__init__(value)
self.value = value
self.max_length = max_length
[docs] def __str__(self) -> str:
return f"unmappable return value length: {len(self.value)} > {self.max_length}"
[docs]class AirflowDagCycleException(AirflowException):
"""Raise when there is a cycle in DAG definition."""
[docs]class AirflowDagDuplicatedIdException(AirflowException):
"""Raise when a DAG's ID is already used by another DAG."""
def __init__(self, dag_id: str, incoming: str, existing: str) -> None:
super().__init__(dag_id, incoming, existing)
self.dag_id = dag_id
self.incoming = incoming
self.existing = existing
[docs] def __str__(self) -> str:
return f"Ignoring DAG {self.dag_id} from {self.incoming} - also found in {self.existing}"
[docs]class AirflowClusterPolicyViolation(AirflowException):
"""Raise when there is a violation of a Cluster Policy in DAG definition."""
[docs]class AirflowClusterPolicySkipDag(AirflowException):
"""Raise when skipping dag is needed in Cluster Policy."""
[docs]class AirflowClusterPolicyError(AirflowException):
"""Raise for a Cluster Policy other than AirflowClusterPolicyViolation or AirflowClusterPolicySkipDag."""
[docs]class DagNotFound(AirflowNotFoundException):
"""Raise when a DAG is not available in the system."""
[docs]class DagCodeNotFound(AirflowNotFoundException):
"""Raise when a DAG code is not available in the system."""
[docs]class DagRunNotFound(AirflowNotFoundException):
"""Raise when a DAG Run is not available in the system."""
[docs]class DagRunAlreadyExists(AirflowBadRequest):
"""Raise when creating a DAG run for DAG which already has DAG run entry."""
def __init__(self, dag_run: DagRun) -> None:
super().__init__(f"A DAG Run already exists for DAG {dag_run.dag_id} with run id {dag_run.run_id}")
self.dag_run = dag_run
[docs] def serialize(self):
cls = self.__class__
# Note the DagRun object will be detached here and fails serialization, we need to create a new one
from airflow.models import DagRun
dag_run = DagRun(
state=self.dag_run.state,
dag_id=self.dag_run.dag_id,
run_id=self.dag_run.run_id,
external_trigger=self.dag_run.external_trigger,
run_type=self.dag_run.run_type,
)
dag_run.id = self.dag_run.id
return (
f"{cls.__module__}.{cls.__name__}",
(),
{"dag_run": dag_run},
)
[docs]class DagFileExists(AirflowBadRequest):
"""Raise when a DAG ID is still in DagBag i.e., DAG file is in DAG folder."""
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
warnings.warn("DagFileExists is deprecated and will be removed.", DeprecationWarning, stacklevel=2)
[docs]class FailStopDagInvalidTriggerRule(AirflowException):
"""Raise when a dag has 'fail_stop' enabled yet has a non-default trigger rule."""
_allowed_rules = (TriggerRule.ALL_SUCCESS, TriggerRule.ALL_DONE_SETUP_SUCCESS)
@classmethod
def check(cls, *, fail_stop: bool, trigger_rule: TriggerRule):
"""
Check that fail_stop dag tasks have allowable trigger rules.
:meta private:
"""
if fail_stop and trigger_rule not in cls._allowed_rules:
raise cls()
[docs] def __str__(self) -> str:
return f"A 'fail-stop' dag can only have {TriggerRule.ALL_SUCCESS} trigger rule"
[docs]class DuplicateTaskIdFound(AirflowException):
"""Raise when a Task with duplicate task_id is defined in the same DAG."""
[docs]class TaskAlreadyInTaskGroup(AirflowException):
"""Raise when a Task cannot be added to a TaskGroup since it already belongs to another TaskGroup."""
def __init__(self, task_id: str, existing_group_id: str | None, new_group_id: str) -> None:
super().__init__(task_id, new_group_id)
self.task_id = task_id
self.existing_group_id = existing_group_id
self.new_group_id = new_group_id
[docs] def __str__(self) -> str:
if self.existing_group_id is None:
existing_group = "the DAG's root group"
else:
existing_group = f"group {self.existing_group_id!r}"
return f"cannot add {self.task_id!r} to {self.new_group_id!r} (already in {existing_group})"
[docs]class SerializationError(AirflowException):
"""A problem occurred when trying to serialize something."""
[docs]class ParamValidationError(AirflowException):
"""Raise when DAG params is invalid."""
[docs]class TaskNotFound(AirflowNotFoundException):
"""Raise when a Task is not available in the system."""
[docs]class TaskInstanceNotFound(AirflowNotFoundException):
"""Raise when a task instance is not available in the system."""
[docs]class PoolNotFound(AirflowNotFoundException):
"""Raise when a Pool is not available in the system."""
[docs]class FileSyntaxError(NamedTuple):
"""Information about a single error in a file."""
[docs] def __str__(self):
return f"{self.message}. Line number: s{str(self.line_no)},"
[docs]class AirflowFileParseException(AirflowException):
"""
Raises when connection or variable file can not be parsed.
:param msg: The human-readable description of the exception
:param file_path: A processed file that contains errors
:param parse_errors: File syntax errors
"""
def __init__(self, msg: str, file_path: str, parse_errors: list[FileSyntaxError]) -> None:
super().__init__(msg)
self.msg = msg
self.file_path = file_path
self.parse_errors = parse_errors
[docs] def __str__(self):
from airflow.utils.code_utils import prepare_code_snippet
from airflow.utils.platform import is_tty
result = f"{self.msg}\nFilename: {self.file_path}\n\n"
for error_no, parse_error in enumerate(self.parse_errors, 1):
result += "=" * 20 + f" Parse error {error_no:3} " + "=" * 20 + "\n"
result += f"{parse_error.message}\n"
if parse_error.line_no:
result += f"Line number: {parse_error.line_no}\n"
if parse_error.line_no and is_tty():
result += "\n" + prepare_code_snippet(self.file_path, parse_error.line_no) + "\n"
return result
[docs]class ConnectionNotUnique(AirflowException):
"""Raise when multiple values are found for the same connection ID."""
[docs]class TaskDeferred(BaseException):
"""
Signal an operator moving to deferred state.
Special exception raised to signal that the operator it was raised from
wishes to defer until a trigger fires. Triggers can send execution back to task or end the task instance
directly. If the trigger should end the task instance itself, ``method_name`` does not matter,
and can be None; otherwise, provide the name of the method that should be used when
resuming execution in the task.
"""
def __init__(
self,
*,
trigger,
method_name: str,
kwargs: dict[str, Any] | None = None,
timeout: timedelta | int | float | None = None,
):
super().__init__()
self.trigger = trigger
self.method_name = method_name
self.kwargs = kwargs
self.timeout: timedelta | None
# Check timeout type at runtime
if isinstance(timeout, (int, float)):
self.timeout = timedelta(seconds=timeout)
else:
self.timeout = timeout
if self.timeout is not None and not hasattr(self.timeout, "total_seconds"):
raise ValueError("Timeout value must be a timedelta")
[docs] def serialize(self):
cls = self.__class__
return (
f"{cls.__module__}.{cls.__name__}",
(),
{
"trigger": self.trigger,
"method_name": self.method_name,
"kwargs": self.kwargs,
"timeout": self.timeout,
},
)
[docs] def __repr__(self) -> str:
return f"<TaskDeferred trigger={self.trigger} method={self.method_name}>"
[docs]class TaskDeferralError(AirflowException):
"""Raised when a task failed during deferral for some reason."""
[docs]class TaskDeferralTimeout(AirflowException):
"""Raise when there is a timeout on the deferral."""
# The try/except handling is needed after we moved all k8s classes to cncf.kubernetes provider
# These two exceptions are used internally by Kubernetes Executor but also by PodGenerator, so we need
# to leave them here in case older version of cncf.kubernetes provider is used to run KubernetesPodOperator
# and it raises one of those exceptions. The code should be backwards compatible even if you import
# and try/except the exception using direct imports from airflow.exceptions.
# 1) if you have old provider, both provider and pod generator will throw the "airflow.exceptions" exception.
# 2) if you have new provider, both provider and pod generator will throw the
# "airflow.providers.cncf.kubernetes" as it will be imported here from the provider.
try:
from airflow.providers.cncf.kubernetes.pod_generator import PodMutationHookException
except ImportError:
[docs] class PodMutationHookException(AirflowException): # type: ignore[no-redef]
"""Raised when exception happens during Pod Mutation Hook execution."""
try:
from airflow.providers.cncf.kubernetes.pod_generator import PodReconciliationError
except ImportError:
[docs] class PodReconciliationError(AirflowException): # type: ignore[no-redef]
"""Raised when an error is encountered while trying to merge pod configs."""
[docs]class RemovedInAirflow3Warning(DeprecationWarning):
"""Issued for usage of deprecated features that will be removed in Airflow3."""
[docs] deprecated_since: str | None = None
"Indicates the airflow version that started raising this deprecation warning"
[docs]class AirflowProviderDeprecationWarning(DeprecationWarning):
"""Issued for usage of deprecated features of Airflow provider."""
[docs] deprecated_provider_since: str | None = None
"Indicates the provider version that started raising this deprecation warning"
[docs]class DeserializingResultError(ValueError):
"""Raised when an error is encountered while a pickling library deserializes a pickle file."""
[docs] def __str__(self):
return (
"Error deserializing result. Note that result deserialization "
"is not supported across major Python versions. Cause: " + str(self.__cause__)
)
[docs]class UnknownExecutorException(ValueError):
"""Raised when an attempt is made to load an executor which is not configured."""