tally
tally copied to clipboard
A Go metrics interface with fast buffered metrics and third party reporters
In file `stats.go`, all code paths accessing `(*counter).prev` are protected with a mutex, yet every time the field is being accessed with an atomic Load/Store operation. These are the possible...
I need to emit latency metrics through m3. Is there a way to emit timer metrics or histogram metrics with weight? Basically, we are emitting how long a single request...
Prometheus CachedReporter provides a handler for metrics endpoint. It is useful to have the ability to access baseReporter from Scope so that the handler method can be called from the...
Its currently not possible to send metrics to different reporters if those are not all Reporters or either CachedReports. This does not work, as it will report values only to...
initial support for datadog service. to test it with a live account, run ``` API_KEY=your-datdog-key go test -v ```
Presently, scope and metric caches exist indefinitely. This creates a significant problem for applications dealing with a large number of metrics, particularly if a substantial amount of metrics are emitted...
## Summary | PR Status | Type | Impact level | | :---: | :---: | :---: | | Ready | Feature Request #151 | Medium | ## Description This...
This is somewhat a duplicate of #110. I know that StatsD Reporter doesn't support tagging: https://github.com/uber-go/tally/blob/master/statsd/reporter.go#L151 I believe this is the case because the cactus/go-statsd-client didn't support tags until this...
You can use ReportableScope if you need to send data after some event happened. For example, you can use this interface in HTTP Handler for pull-based metrics model.
Ported from https://github.com/m3db/m3x/pull/186