cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Difference between extracted dataPoints for one timeslot of timeseriesId=com.dynatrace.builtin:pgi.workerprocesses

suresh230591
Advisor

@Wolfgang Beer

We have seen a difference between extracted dataPoints for timeseriesId com.dynatrace.builtin:pgi.workerprocesses for a single time slot. It is "null" in the first call and "0" in the second.What could be the reason for this?

For example: We extracted data for last one hour with one minute resolution

At time 19:00 PM

{"result":{"dataPoints":{"PROCESS_GROUP_INSTANCE-6E9832E802797A95":[[1539712800000,0.0],[1539712860000,0.0],[1539712920000,0.0],[1539712980000,0.0],[1539713040000,0.0],[1539713100000,0.0],[1539713160000,0.0],[1539713220000,0.0],[1539713280000,0.0],[1539713340000,0.0],[1539713400000,0.0],[1539713460000,0.0],[1539713520000,0.0],[1539713580000,0.0],[1539713640000,0.0],[1539713700000,0.0],[1539713760000,0.0],[1539713820000,1.0],[1539713880000,1.0],[1539713940000,0.0],[1539714000000,0.0],[1539714060000,0.0],[1539714120000,0.0],[1539714180000,0.0],[1539714240000,0.0],[1539714300000,0.0],[1539714360000,0.0],[1539714420000,0.0],[1539714480000,0.0],[1539714540000,0.0],[1539714600000,0.0],[1539714660000,0.0],[1539714720000,0.0],[1539714780000,0.0],[1539714840000,0.0],[1539714900000,0.0],[1539714960000,0.0],[1539715020000,0.0],[1539715080000,0.0],[1539715140000,0.0],[1539715200000,0.0],[1539715260000,0.0],[1539715320000,0.0],[1539715380000,0.0],[1539715440000,0.0],[1539715500000,0.0],[1539715560000,0.0],[1539715620000,0.0],[1539715680000,0.0],[1539715740000,0.0],[1539715800000,0.0],[1539715860000,0.0],[1539715920000,0.0],[1539715980000,0.0],[1539716040000,0.0],[1539716100000,0.0],[1539716160000,0.0],[1539716220000,0.0],[1539716280000,0.0],[1539716340000,0.0],[1539716400000,null]],

At time 20:00 PM

{"result":{"dataPoints":{"PROCESS_GROUP_INSTANCE-6E9832E802797A95":[[1539716400000,0.0],[1539716460000,0.0],[1539716520000,0.0],[1539716580000,0.0],[1539716640000,0.0],[1539716700000,0.0],[1539716760000,0.0],[1539716820000,0.0],[1539716880000,0.0],[1539716940000,0.0],[1539717000000,0.0],[1539717060000,0.0],[1539717120000,0.0],[1539717180000,0.0],[1539717240000,0.0],[1539717300000,0.0],[1539717360000,0.0],[1539717420000,0.0],[1539717480000,1.0],[1539717540000,1.0],[1539717600000,0.0],[1539717660000,0.0],[1539717720000,0.0],[1539717780000,0.0],[1539717840000,0.0],[1539717900000,0.0],[1539717960000,0.0],[1539718020000,0.0],[1539718080000,0.0],[1539718140000,0.0],[1539718200000,0.0],[1539718260000,0.0],[1539718320000,0.0],[1539718380000,0.0],[1539718440000,0.0],[1539718500000,0.0],[1539718560000,0.0],[1539718620000,0.0],[1539718680000,0.0],[1539718740000,0.0],[1539718800000,0.0],[1539718860000,0.0],[1539718920000,0.0],[1539718980000,0.0],[1539719040000,0.0],[1539719100000,0.0],[1539719160000,0.0],[1539719220000,0.0],[1539719280000,0.0],[1539719340000,0.0],[1539719400000,0.0],[1539719460000,0.0],[1539719520000,0.0],[1539719580000,0.0],[1539719640000,0.0],[1539719700000,0.0],[1539719760000,0.0],[1539719820000,0.0],[1539719880000,0.0],[1539719940000,null]

In first call at time [1539716400000,null]

In second call at time [1539716400000,0.0]

Thanks

Suresh


4 REPLIES 4

wolfgang_beer
Dynatrace Champion
Dynatrace Champion

The reason for that is that measurements can be late. So in first all the last minute was not written/received from the agent side, while in second call it was already written. We will fix that in the future by not exporting the last minute that has not been written so far.


@Wolfgang B., Thanks for your response

I have following queries , It would be great if you can clarify

1) I have tested few timeseries from host, service and process entities. Every time result has one extra dataPoint with value "null". Is this the normal behavior and applies to all timeseries.

2) In some cases, even last 2 dataPoints are "null". Does this mean that last 2 minutes data is not reported by agent. Is it always restrict to last 1 or 2 minutes?. See below example

com.dynatrace.builtin:pgi.jvm.threadcount


{"result":{"dataPoints":{"PROCESS_GROUP_INSTANCE-E880491D6D579D66":[[1540355520000,102.0],[1540355580000,102.0],[1540355640000,102.0],[1540355700000,102.0],[1540355760000,null],[1540355820000,null]

com.dynatrace.builtin:service.requests


{"result":{"dataPoints":{"SERVICE-A7786473001CBAED":[[1540356240000,21.0],[1540356300000,36.0],[1540356360000,88.0],[1540356420000,116.0],[1540356480000,null],[1540356540000,null].

The results are inconsistent and dataPoints are not returning the exact values.This will impact our service assurance solution which is collecting this data.

3) When this is going to fixed. Any timeline please

@Wolfgang B Could you please comment on this. This is creating problem to service assurance to aggregate the data.

Thanks

Suresh


I'm seeing the same behaviour extracting by the minute this timeseries:


com.dynatrace.builtin:pgi.workerprocesses


"PROCESS_GROUP_INSTANCE-BE95C36E10F9B54A": [[1588236180000,1],[1588236240000,null]

and in a way it make sense because "1588236180000" is the minute before I made the call (so value is already calculated) and "1588236240000" is the minute while I'm making the call (where you can't say if the value will be one or zero bacause the minute is still not finished?)


I'm trying to give the best explanation to my knowledge but I'm not quite sure it is.


What worries me the most is that if I make this call every minute I see only one istance at the time of a group of six... if the same call is made in a 5 previous-minute I can see all six istances of the group.


What were your conclusion about it @Suresh K.?

suresh230591
Advisor

One more thing we have noticed , If we fetch last one hour data with one minute resolution. It give us 61 dataPoints with last value "null", ideally it should be 60 . Is this the normal behavior.

This is same case for service timeseries where if we fetch 5 min data , it is giving us 6 values with last value "null".

Is the last value always going to be null for all timeseries.