Asaf Mesika
Asaf Mesika
@jack-berg, thanks a lot for answering. First, I agree with you it's a challenging effort and involve work to be done. Since summary has existed for a long time, perhaps...
> there is no OTel API to delete a timeseries, and we are not willing to change this detail before reaching a Metrics-1.0 milestone. @jmacd Can you please explain the...
@dashpole >For users who need a mechanism to report metrics which have a bounded cardinality in a particular collection round, but unbounded cardinality over time (e.g. something like kube-state-metrics which...
> The result would be: > If the instrument matches one or more view's selection criteria, use the aggregation specified by the view. Else if the metric reader (or associated...
One question popped into my mind: Will Hints be typed? If so, will the aggregations allowed be constrained only to the ones specified [here](https://opentelemetry.io/docs/reference/specification/metrics/sdk/#aggregation), i.e., Explicit Bucket and Exponential Bucket?...
>If you're referring to an aggregation that produces [summary type metrics](https://github.com/open-telemetry/opentelemetry-specification/blob/main/specification/metrics/data-model.md#summary-legacy), that wouldn't be possible because there currently is no summary [aggregation](https://github.com/open-telemetry/opentelemetry-specification/blob/main/specification/metrics/data-model.md#summary-legacy). My thinking was creating my own SDK (my...
@jack-berg How can we move your idea forward of amending the Histogram interface to have a way to configure aggregation and its options/configuration?
🥳 Brilliant work @jack-berg
Maybe you accidentally used the same buffer directory concurrently by two processes? Or mixed somehow ? On Mon, 21 Oct 2019 at 16:49 StephanPraetsch wrote: > Additional Information: We fixed...
I think we can easily add a configuration for the field names sent in the JSON, thus it will have a map of key fields like: ``` loglevel message ```...