Sometimes, you might want to backfill or replay old data into your queue. Here's an example of enumerating a day of files and replaying them into the queue so the s3-notification job picks them up again.
Before starting, define some configuration values...
$ export BACKFILL_QUEUE_URL="https://sqs.us-east-1.amazonaws.com/${AWS_ACCOUNT_ID}/l4aws-cloudtrail"
$ export BACKFILL_BUCKET="example-aws-logs-us-east-1"
$ export BACKFILL_DATE="2015/01/01"
Next, you'll want to get the list of files you'll be replaying...
$ aws s3api list-objects \
--bucket="${BACKFILL_BUCKET}" \
--prefix="AWS/CloudTrail/AWSLogs/${AWS_ACCOUNT_ID}/CloudTrail/${AWS_DEFAULT_REGION}/${BACKFILL_DATE}" \
> tmp-replay-$$.json
Then, cat the results into the appropriate helper script based on the job you're using...
$ cat tmp-replay-$$.json | ./bin/s3-notification
Once finished, you can remove the temporary results...
$ rm tmp-replay-$$.json
It may take some time before everything finishes processing.