Github user revans2 commented on the issue:
https://github.com/apache/storm/pull/2377
@BigOneLiu actually like I sad they are **not** coming from the same
location. 99.9% of the time they are probably going to be the same underlying
value, but if for some reason the metrics tuple is delayed by 1 second they are
100% guaranteed to be different.
@vesense if you are concerned about this that is probably a good enough
reason to not do it. Users are going to have enough on their hands going to
2.x where we may even be moving to an entirely new metrics system.
---
