This library provides a metrics
package which can be used to instrument code,
expose application metrics, and profile runtime performance in a flexible manner.
The metrics
package makes use of a MetricSink
interface to support delivery
to any type of backend. Currently the following sinks are provided:
- StatsiteSink : Sinks to a statsite instance (TCP)
- StatsdSink: Sinks to a StatsD / statsite instance (UDP)
- PrometheusSink: Sinks to a Prometheus metrics endpoint (exposed via HTTP for scrapes)
- InmemSink : Provides in-memory aggregation, can be used to export stats
- FanoutSink : Sinks to multiple sinks. Enables writing to multiple statsite instances for example.
- BlackholeSink : Sinks to nowhere
In addition to the sinks, the InmemSignal
can be used to catch a signal,
and dump a formatted output of recent metrics. For example, when a process gets
a SIGUSR1, it can dump to stderr recent performance metrics for debugging.
Most metrics do have an equivalent ending with WithLabels
, such methods
allow to push metrics with labels and use some features of underlying Sinks
(ex: translated into Prometheus labels).
Since some of these labels may increase the cardinality of metrics, the library allows filtering labels using a allow/block list filtering system which is global to all metrics.
- If
Config.AllowedLabels
is not nil, then only labels specified in this value will be sent to underlying Sink, otherwise, all labels are sent by default. - If
Config.BlockedLabels
is not nil, any label specified in this value will not be sent to underlying Sinks.
By default, both Config.AllowedLabels
and Config.BlockedLabels
are nil, meaning that
no tags are filtered at all, but it allows a user to globally block some tags with high
cardinality at the application level.
v0.5.0 of the library renamed the Go module from github.com/armon/go-metrics
to github.com/hashicorp/go-metrics
.
While this did not introduce any breaking changes to the API, the change did subtly break backwards compatibility.
In essence, Go treats a renamed module as entirely distinct and will happily compile both modules into the same binary.
Due to most uses of the go-metrics library involving emitting metrics via the global metrics handler, having two global
metrics handlers could cause a subset of metrics to be effectively lost. As an example, if your application configures
go-metrics exporting via the armon
namespace, then any metrics sent to go-metrics via the hashicorp
namespaced module
will never get exported.
Eventually all usage of armon/go-metrics
should be replaced with usage of hashicorp/go-metrics
. However, a single
point-in-time coordinated update across all libraries that an application may depend on isn't always feasible. To facilitate migrations,
a github.com/hashicorp/go-metrics/compat
package has been introduced. This package and sub-packages are API compatible with
armon/go-metrics
. Libraries should be updated to use this package for emitting metrics via the global handlers. Internally,
the package will route metrics to either armon/go-metrics
or hashicorp/go-metrics
. This is achieved at a global level
within an application via the use of Go build tags.
Build Tags
armonmetrics
- Using this tag will cause metrics to be routed toarmon/go-metrics
hashicorpmetrics
- Using this tag will cause all metrics to be routed tohashicorp/go-metrics
If no build tag is specified, the default behavior is to use armon/go-metrics
. The overall migration path would be as follows:
- Upgrade libraries using
armon/go-metrics
to consumehashicorp/go-metrics/compat
instead. - Update library dependencies of applications that use
armon/go-metrics
.- This doesn't need to be one big atomic update but can be slower due to the default behavior remaining unaltered.
- At this point all metrics will still be emitted to
armon/go-metrics
- Update the application to use
hashicorp/go-metrics
- Replace all application imports of
github.com/armon/go-metrics
withgithub.com/hashicorp/go-metrics
- Libraries are unaltered at this stage.
- Instrument your build system to build with the
hashicorpmetrics
tag.
- Replace all application imports of
Your migration is effectively finished and your application is now exclusively using hashicorp/go-metrics
. A future release of the library
will change the default behavior to use hashicorp/go-metrics
instead of armon/go-metrics
. At that point in time, any application that
needs more time before performing the migration must instrument their build system to include the armonmetrics
tag. A subsequent release
after that will eventually remove the compatibility layer all together. The rough timeline for this will be mid-2025 for changing the default
behavior and then the end of 2025 for removal of the compatibility layer.
Here is an example of using the package:
func SlowMethod() {
// Profiling the runtime of a method
defer metrics.MeasureSince([]string{"SlowMethod"}, time.Now())
}
// Configure a statsite sink as the global metrics sink
sink, _ := metrics.NewStatsiteSink("statsite:8125")
metrics.NewGlobal(metrics.DefaultConfig("service-name"), sink)
// Emit a Key/Value pair
metrics.EmitKey([]string{"questions", "meaning of life"}, 42)
Here is an example of setting up a signal handler:
// Setup the inmem sink and signal handler
inm := metrics.NewInmemSink(10*time.Second, time.Minute)
sig := metrics.DefaultInmemSignal(inm)
metrics.NewGlobal(metrics.DefaultConfig("service-name"), inm)
// Run some code
inm.SetGauge([]string{"foo"}, 42)
inm.EmitKey([]string{"bar"}, 30)
inm.IncrCounter([]string{"baz"}, 42)
inm.IncrCounter([]string{"baz"}, 1)
inm.IncrCounter([]string{"baz"}, 80)
inm.AddSample([]string{"method", "wow"}, 42)
inm.AddSample([]string{"method", "wow"}, 100)
inm.AddSample([]string{"method", "wow"}, 22)
....
When a signal comes in, output like the following will be dumped to stderr:
[2014-01-28 14:57:33.04 -0800 PST][G] 'foo': 42.000
[2014-01-28 14:57:33.04 -0800 PST][P] 'bar': 30.000
[2014-01-28 14:57:33.04 -0800 PST][C] 'baz': Count: 3 Min: 1.000 Mean: 41.000 Max: 80.000 Stddev: 39.509
[2014-01-28 14:57:33.04 -0800 PST][S] 'method.wow': Count: 3 Min: 22.000 Mean: 54.667 Max: 100.000 Stddev: 40.513