Determine points of message loss

[data-sync-user]

Once a good understanding of message flow through Autopush is combined with learnings from the Analytic Header, we should be able to identify potential points of message loss, which should explain the poor performance of things like Send Tab.

We should create a plan of action to address these to make Push a highly reliable platform that can be used to develop future offerings.

This ticket should include:

• A document outlining how the “stop points” (known milestones a message last traversed) are recorded and collected.
• How those stop points will be collected from various servers and rolled up for reporting.
• Identifying “known” points of loss (e.g. TTL expiration while in storage) vs. unexpected

┆Issue is synchronized with this Jira Task