How to resolve a list of "eventIds" into event attributes?

1

I have worked out how to gather a list of Event Ids, but cannot work out how to use them.

When my glue workflow is triggered (by a change to an S3 bucket), I was hoping to gather some parameters, or attributes, from the triggering event, in order to make decisions about what the workflow should do. Is this possible?

Other than to dedupe, what use is it to have the eventId?

glue_client = boto3.client('glue')
workflow_args = glue_client.get_workflow_run_properties(
    Name=args['WORKFLOW_NAME'], RunId=args['WORKFLOW_RUN_ID']
)["RunProperties"]
logger.info(pprint.pformat(workflow_args))
#  {'aws:eventIds': '[bc432a05-078e-21d8-07c4-2bc86175c476]'}
profile picture
已提问 1 年前271 查看次数
1 回答
1

Hi, From your post, I understand that you are trying to design a use case, whenever there is a s3 event a glue Workflow is triggered and you are looking for options to get other details of the event id key. The "get_workflow_run_properties" API call will only return "aws:eventIds" within response element. This "aws:eventIds" matches with the "id" field returned from the "lookup_events" Cloudtrail API call.

As per the document [1] if you are using the eventbridge event to achieve your use case, please note that Eventbridge emits "NotifyEvent" event when it triggers the Glue workflow. You could look for Glue NotifyEvent API calls in CloudTrail to fetch the entire event that corresponds to the event id of the workflow run.

Sample code:

            glue_client = boto3.client("glue")
            args = getResolvedOptions(sys.argv, ['JOB_NAME','WORKFLOW_NAME', 'WORKFLOW_RUN_ID'])
            workflow_name = args['WORKFLOW_NAME']
            workflow_run_id = args['WORKFLOW_RUN_ID']

            workflow_params = glue_client.get_workflow_run_properties(Name=workflow_name,
                                                  RunId=workflow_run_id)["RunProperties"]
            batched_events = workflow_params['aws:eventIds']

            cloudtrail_client = boto3.client('cloudtrail', region_name='us-east-1')

            response = cloudtrail_client.lookup_events(
                LookupAttributes=[
                    {
                        'AttributeKey': 'EventName',
                        'AttributeValue': 'NotifyEvent'

                    },
                ],
                StartTime=(datetime.now  () - timedelta(minutes=10)),
                EndTime=datetime.now(),
                MaxResults=100
            )

            events = response.get("Events",[])

            for event in events:
            	cloudtrail_event = event['CloudTrailEvent']
                event_payload=json.loads(cloudtrail_event)['requestParameters']['eventPayload']
                if "[{}]".format(event_payload['eventId']) == batched_events:
                    print ("Details :: ",event_payload['eventBody'])

As a workaround you can configure a lambda function as EventBridge rule target (instead of setting Glue workflow as target) and then invoke Glue workflow or job from lambda function and pass the event details as workflow run properties or job parameters. Reference: [1] https://docs.aws.amazon.com/glue/latest/dg/workflows_overview.html [2] https://aws.amazon.com/blogs/big-data/build-a-serverless-event-driven-workflow-with-aws-glue-and-amazon-eventbridge/
[3] https://docs.aws.amazon.com/glue/latest/dg/workflows_overview.html
[4] https://docs.aws.amazon.com/glue/latest/dg/starting-workflow-eventbridge.html

AWS
支持工程师
已回答 1 年前

您未登录。 登录 发布回答。

一个好的回答可以清楚地解答问题和提供建设性反馈,并能促进提问者的职业发展。

回答问题的准则