In the time restricted updates policy, we want to be able to measure the impact of the policy on update frequency, and whether the policy is making devices fall behind schedule. Having metrics such as the time stamps in which updates were blocked by our policy and the amount of times the policy blocked updates would let us quantify the impact on fleets. This would allow us to know if the security concern raised on the policy's design doc (go/cros-update-time-restrictions-dd
) is valid, and we are delaying updates for too long.

The aforementioned feature can not be implemented because update engine currently doesn't expose any policy related metrics. Adding support for this through the update engine's metrics reporter would make it easier for us to implement the feature. This would also be beneficial for future or current policies, since they would also be able to do a similar analysis for their policy's usage.