Fork me on GitHub

Yes, I think so similar. Create instances with rsync /S3, and deploy to s3. With that ASG/Load balancer you can setup triggers and np. You can sync both way s3 (eg just SQLite things, cloudfront dir) or use db to store things


I think, you didn’t use rsync —-delete or similar command plus a “cron job”


But I don’t know that is helpful or not for you.


Has anybody had experience with appending to S3 files continuously? I see one would need to read + append + write and I was wondering how slow that can be. Of course it depends on the payload and the goal is to split in sizeable ones, just was interested in understanding what are baseline numbers...