Hi Ivan
The goal of the archiver is to take older messages, runs, etc, out of the database, and into an object store like s3. This ensures that the database size is bounded and doesn't grow too big, which would affect query performance and cost (object store storage is a lot cheaper than the more performant database storage). It's configurable as to how much history you want to leave in the database.
The effect of this is that once you go passed a certain point in history, it will no longer show in the normal places in the UI and API, instead you'll have to download the archived files if you want access to that old data (and there is a UI to be able to do that).
Regards,
Rudi.