I want to enable cloudtrail logs for my account and so need to create an s3 bucket.I wanted to automate this task using Boto3.Currently I am using the following script
First, in boto3, if you setup security using "aws configure" , you don't need to declare that "sess" section (http://docs.aws.amazon.com/cli/latest/userguide/cli-chap-getting-started.html)
# if you already done aws configure
import boto3
s3 = boto3.client("s3")
s3.create_bucket(Bucket="mybucket", ....)
Second, is the bad boto3 documentation that fail to link proper information. This is found under boto3 pdf, page 2181 (https://media.readthedocs.org/pdf/boto3/latest/boto3.pdf)
Email : The value in the Grantee object is the registered email address of an AWS account.
Grantee : The AWS user or group that you want to have access to transcoded files and playlists. To identify the user or group, you can specify the canonical user ID for an AWS account, an origin access identity for a CloudFront distribution, the registered email address of an AWS account, or a predefined Amazon S3 group
And the easier solution is just use policy setting (http://support.cloudcheckr.com/getting-started-with-cloudcheckr/preparing-your-aws-account/aggregate-cloudtrail/) . You can convert the whole stuff using put_bucket_policy(), skip the dire GrantWrite,GrantWriteACP