#
# 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.
from __future__ import annotations
import asyncio
import time
import boto3
from botocore.exceptions import ClientError
from airflow.exceptions import AirflowException
from airflow.providers.amazon.aws.hooks.base_aws import AwsBaseHook
[docs]DEFAULT_LOG_SUFFIX = "output"
[docs]ERROR_LOG_SUFFIX = "error"
[docs]class GlueJobHook(AwsBaseHook):
"""
Interact with AWS Glue.
Provide thick wrapper around :external+boto3:py:class:`boto3.client("glue") <Glue.Client>`.
:param s3_bucket: S3 bucket where logs and local etl script will be uploaded
:param job_name: unique job name per AWS account
:param desc: job description
:param concurrent_run_limit: The maximum number of concurrent runs allowed for a job
:param script_location: path to etl script on s3
:param retry_limit: Maximum number of times to retry this job if it fails
:param num_of_dpus: Number of AWS Glue DPUs to allocate to this Job
:param region_name: aws region name (example: us-east-1)
:param iam_role_name: AWS IAM Role for Glue Job Execution. If set `iam_role_arn` must equal None.
:param iam_role_arn: AWS IAM Role ARN for Glue Job Execution, If set `iam_role_name` must equal None.
:param create_job_kwargs: Extra arguments for Glue Job Creation
:param update_config: Update job configuration on Glue (default: False)
Additional arguments (such as ``aws_conn_id``) may be specified and
are passed down to the underlying AwsBaseHook.
.. seealso::
- :class:`airflow.providers.amazon.aws.hooks.base_aws.AwsBaseHook`
"""
[docs] class LogContinuationTokens:
"""Used to hold the continuation tokens when reading logs from both streams Glue Jobs write to."""
def __init__(self):
self.output_stream_continuation: str | None = None
self.error_stream_continuation: str | None = None
def __init__(
self,
s3_bucket: str | None = None,
job_name: str | None = None,
desc: str | None = None,
concurrent_run_limit: int = 1,
script_location: str | None = None,
retry_limit: int = 0,
num_of_dpus: int | float | None = None,
iam_role_name: str | None = None,
iam_role_arn: str | None = None,
create_job_kwargs: dict | None = None,
update_config: bool = False,
job_poll_interval: int | float = 6,
*args,
**kwargs,
):
self.job_name = job_name
self.desc = desc
self.concurrent_run_limit = concurrent_run_limit
self.script_location = script_location
self.retry_limit = retry_limit
self.s3_bucket = s3_bucket
self.role_name = iam_role_name
self.role_arn = iam_role_arn
self.s3_glue_logs = "logs/glue-logs/"
self.create_job_kwargs = create_job_kwargs or {}
self.update_config = update_config
self.job_poll_interval = job_poll_interval
worker_type_exists = "WorkerType" in self.create_job_kwargs
num_workers_exists = "NumberOfWorkers" in self.create_job_kwargs
if self.role_arn and self.role_name:
raise ValueError("Cannot set iam_role_arn and iam_role_name simultaneously")
if worker_type_exists and num_workers_exists:
if num_of_dpus is not None:
raise ValueError("Cannot specify num_of_dpus with custom WorkerType")
elif not worker_type_exists and num_workers_exists:
raise ValueError("Need to specify custom WorkerType when specifying NumberOfWorkers")
elif worker_type_exists and not num_workers_exists:
raise ValueError("Need to specify NumberOfWorkers when specifying custom WorkerType")
elif num_of_dpus is None:
self.num_of_dpus: int | float = 10
else:
self.num_of_dpus = num_of_dpus
kwargs["client_type"] = "glue"
super().__init__(*args, **kwargs)
[docs] def create_glue_job_config(self) -> dict:
default_command = {
"Name": "glueetl",
"ScriptLocation": self.script_location,
}
command = self.create_job_kwargs.pop("Command", default_command)
if not self.role_arn:
execution_role = self.get_iam_execution_role()
role_arn = execution_role["Role"]["Arn"]
else:
role_arn = self.role_arn
config = {
"Name": self.job_name,
"Description": self.desc,
"Role": role_arn,
"ExecutionProperty": {"MaxConcurrentRuns": self.concurrent_run_limit},
"Command": command,
"MaxRetries": self.retry_limit,
**self.create_job_kwargs,
}
if hasattr(self, "num_of_dpus"):
config["MaxCapacity"] = self.num_of_dpus
if self.s3_bucket is not None:
config["LogUri"] = f"s3://{self.s3_bucket}/{self.s3_glue_logs}{self.job_name}"
return config
[docs] def list_jobs(self) -> list:
"""
Get list of Jobs.
.. seealso::
- :external+boto3:py:meth:`Glue.Client.get_jobs`
"""
return self.conn.get_jobs()
[docs] def get_iam_execution_role(self) -> dict:
try:
iam_client = self.get_session(region_name=self.region_name).client(
"iam", endpoint_url=self.conn_config.endpoint_url, config=self.config, verify=self.verify
)
glue_execution_role = iam_client.get_role(RoleName=self.role_name)
self.log.info("Iam Role Name: %s", self.role_name)
return glue_execution_role
except Exception as general_error:
self.log.error("Failed to create aws glue job, error: %s", general_error)
raise
[docs] def initialize_job(
self,
script_arguments: dict | None = None,
run_kwargs: dict | None = None,
) -> dict[str, str]:
"""
Initialize connection with AWS Glue to run job.
.. seealso::
- :external+boto3:py:meth:`Glue.Client.start_job_run`
"""
script_arguments = script_arguments or {}
run_kwargs = run_kwargs or {}
try:
if self.update_config:
job_name = self.create_or_update_glue_job()
else:
job_name = self.get_or_create_glue_job()
return self.conn.start_job_run(JobName=job_name, Arguments=script_arguments, **run_kwargs)
except Exception as general_error:
self.log.error("Failed to run aws glue job, error: %s", general_error)
raise
[docs] def get_job_state(self, job_name: str, run_id: str) -> str:
"""
Get state of the Glue job; the job state can be running, finished, failed, stopped or timeout.
.. seealso::
- :external+boto3:py:meth:`Glue.Client.get_job_run`
:param job_name: unique job name per AWS account
:param run_id: The job-run ID of the predecessor job run
:return: State of the Glue job
"""
job_run = self.conn.get_job_run(JobName=job_name, RunId=run_id, PredecessorsIncluded=True)
return job_run["JobRun"]["JobRunState"]
[docs] async def async_get_job_state(self, job_name: str, run_id: str) -> str:
"""
Get state of the Glue job; the job state can be running, finished, failed, stopped or timeout.
The async version of get_job_state.
"""
async with self.async_conn as client:
job_run = await client.get_job_run(JobName=job_name, RunId=run_id)
return job_run["JobRun"]["JobRunState"]
[docs] def print_job_logs(
self,
job_name: str,
run_id: str,
continuation_tokens: LogContinuationTokens,
):
"""
Print the latest job logs to the Airflow task log and updates the continuation tokens.
:param continuation_tokens: the tokens where to resume from when reading logs.
The object gets updated with the new tokens by this method.
"""
log_client = boto3.client("logs")
paginator = log_client.get_paginator("filter_log_events")
def display_logs_from(log_group: str, continuation_token: str | None) -> str | None:
"""Mutualize iteration over the 2 different log streams glue jobs write to."""
fetched_logs = []
next_token = continuation_token
try:
for response in paginator.paginate(
logGroupName=log_group,
logStreamNames=[run_id],
PaginationConfig={"StartingToken": continuation_token},
):
fetched_logs.extend([event["message"] for event in response["events"]])
# if the response is empty there is no nextToken in it
next_token = response.get("nextToken") or next_token
except ClientError as e:
if e.response["Error"]["Code"] == "ResourceNotFoundException":
# we land here when the log groups/streams don't exist yet
self.log.warning(
"No new Glue driver logs so far.\nIf this persists, check the CloudWatch dashboard "
f"at: https://{self.conn_region_name}.console.aws.amazon.com/cloudwatch/home"
)
else:
raise
if len(fetched_logs):
# Add a tab to indent those logs and distinguish them from airflow logs.
# Log lines returned already contain a newline character at the end.
messages = "\t".join(fetched_logs)
self.log.info("Glue Job Run %s Logs:\n\t%s", log_group, messages)
else:
self.log.info("No new log from the Glue Job in %s", log_group)
return next_token
log_group_prefix = self.conn.get_job_run(JobName=job_name, RunId=run_id)["JobRun"]["LogGroupName"]
log_group_default = f"{log_group_prefix}/{DEFAULT_LOG_SUFFIX}"
log_group_error = f"{log_group_prefix}/{ERROR_LOG_SUFFIX}"
# one would think that the error log group would contain only errors, but it actually contains
# a lot of interesting logs too, so it's valuable to have both
continuation_tokens.output_stream_continuation = display_logs_from(
log_group_default, continuation_tokens.output_stream_continuation
)
continuation_tokens.error_stream_continuation = display_logs_from(
log_group_error, continuation_tokens.error_stream_continuation
)
[docs] def job_completion(self, job_name: str, run_id: str, verbose: bool = False) -> dict[str, str]:
"""
Wait until Glue job with job_name finishes; return final state if finished or raises AirflowException.
:param job_name: unique job name per AWS account
:param run_id: The job-run ID of the predecessor job run
:param verbose: If True, more Glue Job Run logs show in the Airflow Task Logs. (default: False)
:return: Dict of JobRunState and JobRunId
"""
next_log_tokens = self.LogContinuationTokens()
while True:
job_run_state = self.get_job_state(job_name, run_id)
ret = self._handle_state(job_run_state, job_name, run_id, verbose, next_log_tokens)
if ret:
return ret
else:
time.sleep(self.job_poll_interval)
[docs] async def async_job_completion(self, job_name: str, run_id: str, verbose: bool = False) -> dict[str, str]:
"""
Wait until Glue job with job_name finishes; return final state if finished or raises AirflowException.
:param job_name: unique job name per AWS account
:param run_id: The job-run ID of the predecessor job run
:param verbose: If True, more Glue Job Run logs show in the Airflow Task Logs. (default: False)
:return: Dict of JobRunState and JobRunId
"""
next_log_tokens = self.LogContinuationTokens()
while True:
job_run_state = await self.async_get_job_state(job_name, run_id)
ret = self._handle_state(job_run_state, job_name, run_id, verbose, next_log_tokens)
if ret:
return ret
else:
await asyncio.sleep(self.job_poll_interval)
def _handle_state(
self,
state: str,
job_name: str,
run_id: str,
verbose: bool,
next_log_tokens: GlueJobHook.LogContinuationTokens,
) -> dict | None:
"""Process Glue Job state while polling; used by both sync and async methods."""
failed_states = ["FAILED", "TIMEOUT"]
finished_states = ["SUCCEEDED", "STOPPED"]
if verbose:
self.print_job_logs(
job_name=job_name,
run_id=run_id,
continuation_tokens=next_log_tokens,
)
if state in finished_states:
self.log.info("Exiting Job %s Run State: %s", run_id, state)
return {"JobRunState": state, "JobRunId": run_id}
if state in failed_states:
job_error_message = f"Exiting Job {run_id} Run State: {state}"
self.log.info(job_error_message)
raise AirflowException(job_error_message)
else:
self.log.info(
"Polling for AWS Glue Job %s current run state with status %s",
job_name,
state,
)
return None
[docs] def has_job(self, job_name) -> bool:
"""
Check if the job already exists.
.. seealso::
- :external+boto3:py:meth:`Glue.Client.get_job`
:param job_name: unique job name per AWS account
:return: Returns True if the job already exists and False if not.
"""
self.log.info("Checking if job already exists: %s", job_name)
try:
self.conn.get_job(JobName=job_name)
return True
except self.conn.exceptions.EntityNotFoundException:
return False
[docs] def update_job(self, **job_kwargs) -> bool:
"""
Update job configurations.
.. seealso::
- :external+boto3:py:meth:`Glue.Client.update_job`
:param job_kwargs: Keyword args that define the configurations used for the job
:return: True if job was updated and false otherwise
"""
job_name = job_kwargs.pop("Name")
current_job = self.conn.get_job(JobName=job_name)["Job"]
update_config = {
key: value for key, value in job_kwargs.items() if current_job.get(key) != job_kwargs[key]
}
if update_config != {}:
self.log.info("Updating job: %s", job_name)
self.conn.update_job(JobName=job_name, JobUpdate=job_kwargs)
self.log.info("Updated configurations: %s", update_config)
return True
else:
return False
[docs] def get_or_create_glue_job(self) -> str | None:
"""
Get (or creates) and returns the Job name.
.. seealso::
- :external+boto3:py:meth:`Glue.Client.create_job`
:return:Name of the Job
"""
if self.has_job(self.job_name):
return self.job_name
config = self.create_glue_job_config()
self.log.info("Creating job: %s", self.job_name)
self.conn.create_job(**config)
return self.job_name
[docs] def create_or_update_glue_job(self) -> str | None:
"""
Create (or update) and return the Job name.
.. seealso::
- :external+boto3:py:meth:`Glue.Client.update_job`
- :external+boto3:py:meth:`Glue.Client.create_job`
:return:Name of the Job
"""
config = self.create_glue_job_config()
if self.has_job(self.job_name):
self.update_job(**config)
else:
self.log.info("Creating job: %s", self.job_name)
self.conn.create_job(**config)
return self.job_name
