Update buildkite-metrics to use the agent metrics api

.buildkite/steps/build-lambda.sh

@@ -6,8 +6,6 @@ go_src_dir="/go/src/${go_pkg}"
 version=$(awk -F\" '/const Version/ {print $2}' version/version.go)
 dist_file="dist/buildkite-metrics-v${version}-lambda.zip"
 
-docker run --rm -v "${PWD}:${go_src_dir}" -w "${go_src_dir}" eawsy/aws-lambda-go
 mkdir -p dist/
-mv handler.zip "$dist_file"
-
+docker run --rm -v "${PWD}:${go_src_dir}" -w "${go_src_dir}" eawsy/aws-lambda-go --package "$dist_file"
 buildkite-agent artifact upload "$dist_file"

.buildkite/steps/test.sh

@@ -4,8 +4,8 @@ set -eu
 go_version="1.9.2"
 go_pkg="github.com/buildkite/buildkite-metrics"
 
-  docker run \
-    -v "${PWD}:/go/src/${go_pkg}" \
-    -w "/go/src/${go_pkg}" \
-    --rm "golang:${go_version}" \
-    go test -v ./... 
+docker run \
+  -v "${PWD}:/go/src/${go_pkg}" \
+  -w "/go/src/${go_pkg}" \
+  --rm "golang:${go_version}" \
+  go test -v ./... 

README.md

@@ -1,6 +1,6 @@
 # Buildkite Metrics
 
-A command-line tool for collecting [Buildkite](https://buildkite.com/) build/job statistics for external metrics systems. Currently [AWS Cloudwatch](http://aws.amazon.com/cloudwatch/) and [StatsD](https://github.com/etsy/statsd) are supported.
+A command-line tool for collecting [Buildkite](https://buildkite.com/) agent and job statistics for external metrics systems, focusing on enabling auto—scaling. Currently [AWS Cloudwatch](http://aws.amazon.com/cloudwatch/), [StatsD](https://github.com/etsy/statsd) and [Prometheus](https://prometheus.io) are supported.
 
 [![Build status](https://badge.buildkite.com/80d04fcde3a306bef44e77aadb1f1ffdc20ebb3c8f1f585a60.svg)](https://buildkite.com/buildkite/buildkite-metrics)
 
@@ -12,60 +12,73 @@ Either download the latest binary from [Github Releases](https://github.com/buil
 go get github.com/buildkite/buildkite-metrics
 ```
 
+## Running
+
+Run the metrics agent using an Agent Registration Token, found on the [Buildkite Agents page](https://buildkite.com/organizations/-/agents):
+
+```
+buildkite-metrics -token abc123...
+```
+
+By default this will publish metrics for every queue with connected agents and unfinished jobs, and a total of all metrics. Restrict it to a single queue with `-queue` if you're scaling a single cluster of agents:
+
+```
+buildkite-metrics -token abc123... -queue my-queue
+```
+
 ### Backends
 
 By default metrics will be submitted to CloudWatch but the backend can be switched to StatsD or Prometheus using the command-line argument `-backend statsd` or `-backend prometheus` respectively.
 
 The StatsD backend supports the following arguments
 
 * `-statsd-host HOST`: The StatsD host and port (defaults to `127.0.0.1:8125`).
-* `-statsd-tags`: Some StatsD servers like the agent provided by DataDog support tags. If specified, metrics will be tagged by `queue` and `pipeline` otherwise metrics will include the queue/pipeline name in the metric. Only enable this option if you know your StatsD server supports tags.
+* `-statsd-tags`: Some StatsD servers like the agent provided by DataDog support tags. If specified, metrics will be tagged by `queue` otherwise metrics will include the queue name in the metric. Only enable this option if you know your StatsD server supports tags.
 
 The Prometheus backend supports the following arguments
 
 * `-prometheus-addr`: The local address to listen on (defaults to `:8080`).
 * `-prometheus-path`: The path under `prometheus-addr` to expose metrics on (defaults to `/metrics`).
 
+### Upgrading from v2
+
+1. The `-org` argument is no longer needed
+2. The `-token` argument is now an _Agent Registration Token_ — the same used in the Buildkite Agent configuration file, and found on the [Buildkite Agents page](https://buildkite.com/organizations/-/agents).
+3. Build and pipeline metrics have been removed, focusing on agents and jobs by queue for auto–scaling.  
+   If you have a compelling reason to gather build or pipeline metrics please continue to use the [previous version](https://github.com/buildkite/buildkite-metrics/releases/tag/v2.1.0) or [open an issue](https://github.com/buildkite/buildkite-metrics/issues) with details.
+
 ## Development
 
 You can build and run the binary tool locally with golang installed:
 
 ```
-go run *.go -org [myorg] -token [buildkite api access token]
+go run *.go -token [buildkite agent registration token]
 ```
 
-Currently this will publish metrics to Cloudwatch under the custom metric prefix of `Buildkite`, using AWS credentials from your environment. The machine will require the [`cloudwatch:PutMetricData`](https://docs.aws.amazon.com/AmazonCloudWatch/latest/DeveloperGuide/publishingMetrics.html) IAM permission, and the Buildkite API Access token requires the scopes `read_pipelines`, `read_builds` and `read_agents`.
+Currently this will publish metrics to Cloudwatch under the custom metric prefix of `Buildkite`, using AWS credentials from your environment. The machine will require the [`cloudwatch:PutMetricData`](https://docs.aws.amazon.com/AmazonCloudWatch/latest/DeveloperGuide/publishingMetrics.html) IAM permission.
 
 ## Metrics
 
-The following metrics are gathered:
+The following metrics are gathered when no specific queue is supplied:
 
 ```
-Buildkite > RunningBuildsCount
 Buildkite > RunningJobsCount
-Buildkite > ScheduledBuildsCount
 Buildkite > ScheduledJobsCount
 Buildkite > UnfinishedJobsCount
 Buildkite > IdleAgentsCount
 Buildkite > BusyAgentsCount
 Buildkite > TotalAgentsCount
 
-Buildkite > (Queue) > RunningBuildsCount
 Buildkite > (Queue) > RunningJobsCount
-Buildkite > (Queue) > ScheduledBuildsCount
 Buildkite > (Queue) > ScheduledJobsCount
 Buildkite > (Queue) > UnfinishedJobsCount
 Buildkite > (Queue) > IdleAgentsCount
 Buildkite > (Queue) > BusyAgentsCount
 Buildkite > (Queue) > TotalAgentsCount
-
-Buildkite > (Pipeline) > RunningBuildsCount
-Buildkite > (Pipeline) > RunningJobsCount
-Buildkite > (Pipeline) > ScheduledBuildsCount
-Buildkite > (Pipeline) > ScheduledJobsCount
-Buildkite > (Pipeline) > UnfinishedJobsCount
 ```
 
+When a queue is specified, only that queue's metrics are published.
+
 ## AWS Lambda
 
 An AWS Lambda bundle is created and published as part of the build process.
@@ -78,7 +91,7 @@ It's entrypoint is `handler.handle`, it requires a `python2.7` environment and m
  - BUILDKITE_QUEUE
  - BUILDKITE_QUIET
 
-Checkout https://github.com/buildkite/elastic-ci-stack-for-aws/blob/master/templates/metrics.yml for examples of usage.
+Take a look at https://github.com/buildkite/elastic-ci-stack-for-aws/blob/master/templates/metrics.yml for examples of usage.
 
 ## License
 

backend/cloudwatch.go

@@ -29,12 +29,6 @@ func (cb *CloudWatchBackend) Collect(r *collector.Result) error {
 		})...)
 	}
 
-	for name, c := range r.Pipelines {
-		metrics = append(metrics, cloudwatchMetrics(c, []*cloudwatch.Dimension{
-			{Name: aws.String("Pipeline"), Value: aws.String(name)},
-		})...)
-	}
-
 	log.Printf("Extracted %d cloudwatch metrics from results", len(metrics))
 
 	for _, chunk := range chunkCloudwatchMetrics(10, metrics) {

backend/prometheus.go

@@ -69,21 +69,6 @@ func (p *Prometheus) Collect(r *collector.Result) error {
 		}
 	}
 
-	for pipeline, counts := range r.Pipelines {
-		for name, value := range counts {
-			gauge, ok := p.pipelines[name]
-			if !ok {
-				gauge = prometheus.NewGaugeVec(prometheus.GaugeOpts{
-					Name: fmt.Sprintf("buildkite_pipelines_%s", camelToUnderscore(name)),
-					Help: fmt.Sprintf("Buildkite Pipelines: %s", name),
-				}, []string{"pipeline"})
-				prometheus.MustRegister(gauge)
-				p.pipelines[name] = gauge
-			}
-			gauge.WithLabelValues(pipeline).Set(float64(value))
-		}
-	}
-
 	return nil
 }
 

backend/prometheus_test.go

@@ -23,32 +23,23 @@ const (
 
 func newTestResult(t *testing.T) *collector.Result {
 	t.Helper()
-	pipelines := map[string]int{
+	totals := map[string]int{
 		"RunningBuildsCount":   runningBuildsCount,
 		"ScheduledBuildsCount": scheduledBuildsCount,
 		"RunningJobsCount":     runningJobsCount,
 		"ScheduledJobsCount":   scheduledJobsCount,
 		"UnfinishedJobsCount":  unfinishedJobsCount,
+		"IdleAgentCount":       idleAgentCount,
+		"BusyAgentCount":       busyAgentCount,
+		"TotalAgentCount":      totalAgentCount,
 	}
 
-	totals := make(map[string]int)
-	for k, v := range pipelines {
-		totals[k] = v
-	}
-	totals["IdleAgentCount"] = idleAgentCount
-	totals["BusyAgentCount"] = busyAgentCount
-	totals["TotalAgentCount"] = totalAgentCount
-
 	res := &collector.Result{
 		Totals: totals,
 		Queues: map[string]map[string]int{
 			"default": totals,
 			"deploy":  totals,
 		},
-		Pipelines: map[string]map[string]int{
-			"pipeline1": pipelines,
-			"pipeline2": pipelines,
-		},
 	}
 	return res
 }
@@ -81,7 +72,7 @@ func gatherMetrics(t *testing.T) map[string]*dto.MetricFamily {
 func TestCollect(t *testing.T) {
 	mfs := gatherMetrics(t)
 
-	if want, have := 21, len(mfs); want != have {
+	if want, have := 16, len(mfs); want != have {
 		t.Errorf("wanted %d Prometheus metrics, have: %d", want, have)
 	}
 
@@ -125,22 +116,6 @@ func TestCollect(t *testing.T) {
 			idleAgentCount,
 			dto.MetricType_GAUGE,
 		},
-		{
-			"Pipelines",
-			"buildkite_pipelines_running_builds_count",
-			"Buildkite Pipelines: RunningBuildsCount",
-			[]string{"pipeline1", "pipeline2"},
-			runningBuildsCount,
-			dto.MetricType_GAUGE,
-		},
-		{
-			"Pipelines",
-			"buildkite_pipelines_unfinished_jobs_count",
-			"Buildkite Pipelines: UnfinishedJobsCount",
-			[]string{"pipeline1", "pipeline2"},
-			unfinishedJobsCount,
-			dto.MetricType_GAUGE,
-		},
 	}
 
 	for _, tc := range tcs {

backend/statsd.go

@@ -47,22 +47,6 @@ func (cb *StatsD) Collect(r *collector.Result) error {
 		}
 	}
 
-	for pipeline, counts := range r.Pipelines {
-		for name, value := range counts {
-			var finalName string
-			tags := []string{}
-			if cb.tagsSupported {
-				finalName = "pipeline." + name
-				tags = []string{"pipeline:" + pipeline}
-			} else {
-				finalName = "pipeline." + pipeline + "." + name
-			}
-			if err := cb.client.Gauge(finalName, float64(value), tags, 1.0); err != nil {
-				return err
-			}
-		}
-	}
-
 	if err := cb.client.Flush(); err != nil {
 		return err
 	}