Many great points here. I was a developer in agile teams for nearly 8 years before managing one. I usually ask the following in addition to the standard questions:
Knowing what we know now:
1. Are any small estimates now large?
2. Are any large estimates now small?
3. Are the acceptance criteria of the user stories still accurate? Obtainable?
If these aren’t voiced daily you’ll often hear them in the 11th hour of and iteration as excuses as to why features aren’t getting done rather than warnings that they are in jeopardy while there is still time to do something about it. #1 and #2 are often omitted because of estimation guilt. #3 Is a common occurrence where product had too limited a view of a system to write an accurate or obtainable feature and developers were only able to identify this upon deeper inspection once the iteration began.