aws s3 sync used in conjunction with filters somtimes re-downloads unmodified file

[rahul-amaram-vizury]

aws s3 sync s3://test-bucket/rahul-dev/archive/ /tmp/archive/ --exclude '*' --include rahul-dev/archive/Application.jar --debug 2>&1 | grep comparator

So, what I am trying to do is to sync s3://test-bucket/rahul-dev/archive/Application.jar to destination folder only if it has changed. But everytime I run this, it re-downloads with the debug statement file does not exist in destination. Is there any fix for this? Is there a better way for identifying changed files from s3 and then only downloading them?

[jamesls]

This appears to be a bug when using sync with only a portion of the bucket, i.e instead of s3://bucket/ /local you use s3://bucket/path/to/thing /local/path/to/thing. Investigating a fix.

[bounav]

Another example of that bug (or a very similar one) is the following command. It re-downloads everything matching the filters (in other words it doesn't sync but copies):

aws s3 sync s3://bucket /path/to/local/folder --exclude "*.ext" --include "b*"

The intent of the command above is to sync all b*.ext files from the bucket to a local folder.

I tried with --size-only and it copies everything over regardless of whether some files have already been downloaded or not.