Add logging to track basic operational status

[kriswehner]

There's several things we now know need from our experimentation for operational support of the system.

There's a few basics:

• Connecting / connected / etc events when replicating
• Slave monitoring (lag from server if possible in both LSN and time)

Then a few key things we have in our system we need to carry forward:

• Logging the lag of a given client when it connects so we can diagnose field issues with people being behind
• Total LSN synced per request