Fix SumAggregation

[ocelotl]

Fixes #3268

@aabmass this is the PR for the issue I mentioned in the last SIG, PTAL ✌️

[ocelotl]

This is a relevant example.

[ocelotl]

Moving here this comment from @aabmass:

That makes sense 👍 one other question, do we keep the oldest start time that had data in this case? I believe we should. Here's a scenario:

• 1st collection at t1, a delta counter has value 5
• 2nd collection at t2, the counter has None as no observations have been made
• 3rd collection at t3, the counter has value 2.

Does the 3rd collection send the time/start_time as (t1, t3) or (t2, t3)?

[aabmass]

@ocelotl

Does the 3rd collection send the time/start_time as (t1, t3) or (t2, t3)?

Looks like this PR saves the oldest previous time to get 1st option, is that right?

[ocelotl]

@ocelotl

Does the 3rd collection send the time/start_time as (t1, t3) or (t2, t3)?

Looks like this PR saves the oldest previous time to get 1st option, is that right?

Yes, that's right, but TBH, I'm not sure if we should do that. The first option would produce a point whose difference between start and end time is larger than "normal". If a periodic metric reader is used, it can be weird to have a point that is "longer" than a collection cycle, right?

[aabmass]

I think this is the expected behavior, see here

For subsequent points in an unbroken sequence:

• For points with delta aggregation temporality, the StartTimeUnixNano of each point matches the TimeUnixNano of the preceding point

[ocelotl]

I think this is the expected behavior, see here

For subsequent points in an unbroken sequence:

• For points with delta aggregation temporality, the StartTimeUnixNano of each point matches the TimeUnixNano of the preceding point

All right, seems like we are doing the right thing, then. PTAL at the rest of the PR when you have a chance 👍

[ocelotl]

After discussing with @lzchen it looks like we also need to take into consideration the instrument monotonicity in collect.

[ocelotl]

This may help:

Here is one way of choosing the correct instrument:

• I want to count something (by recording a delta value):
  • If the value is monotonically increasing (the delta value is always non-negative) - use a Counter.
  • If the value is NOT monotonically increasing (the delta value can be positive, negative or zero) - use an UpDownCounter.
• I want to record or time something, and the statistics about this thing are likely to be meaningful - use a Histogram.
• I want to measure something (by reporting an absolute value):
  • If the measurement values are non-additive, use an Asynchronous Gauge.
  • If the measurement values are additive:
    • If the value is monotonically increasing - use an Asynchronous Counter.
    • If the value is NOT monotonically increasing - use an Asynchronous UpDownCounter.