Oh, it's the timestamps that are next to Exemplars in text format. Ok, I
forgot about them. Thank you!
Fabian
On Wed, Dec 7, 2022 at 9:59 PM wrote:
> One of them is the sample timestamp right? Or am I missing something?
>
> Essentially a sample is (ts, v). And the ts is the timestamp inside
> m
One of them is the sample timestamp right? Or am I missing something?Essentially a sample is (ts, v). And the ts is the timestamp inside metricpoint? It is confusing though because I would expected created to be next to timestamp. But I guess the current way makes sure only counters, histograms and
Hi,
In OpenMetrics protobuf each metric (i.e. each time series) has an optional
timestamp:
message MetricPoint {
oneof value {
UnknownValue unknown_value = 1;
GaugeValue gauge_value = 2;
CounterValue counter_value = 3;
HistogramValue histogram_value = 4;
StateSetV
Hi,
in OpenMetrics, all metrics (metric == time series) are modelled like this:
message Metric {
repeated Label labels = 1;
repeated MetricPoint metric_points = 2;
}
However, Info metrics have another set of Labels as the value of their
MetricPoint:
message InfoValue {
repeated Labe
> if there was the option to have multiple exemplars for one gauge
That's an interesting thought. Actually, OpenTelemetry metrics have an
optional list of Exemplars for each metric, no matter whether it's a
counter, gauge, histogram, or exponential histogram.
https://github.com/open-telemetry/open
Sorry, unclear phrasing – I mean as the one scraping (whether by hand or as
a Prometheus server), how would I know what the semantics of the value in
the provided exemplar are, whether they are a delta or an absolute value?
Again, I think "it's always an absolute value" makes this least surprising.
In general, what is a foot gun to me can be a rocket shoe to you, so I am
in favor of providing them to those who require them, with clear labeling
of the dangers.
Specifically in this case, it has become more common ("beyond corp", "zero
trust") to use HTTPS over the public internet, combined wit
> At scrape time, how would I know which method was used?
The library keeps a list of Exemplars that are provided at scrape time (in
case of a Gauge the list only has 1 element, because Gauges only have 1
Exemplar).
The question is what to put in the list when inc(), dec(), or set() are
called.
At scrape time, how would I know which method was used? "always the total
value of the gauge" seems like the least surprising choice to me. If users
need to track separate exemplars for increment and decrement, they could
use two counters (basically, a non-native UpDownCounter), which
conceptually
9 matches
Mail list logo
