fix(pyroscope.java): infer service_name label if possible (#6184)

* fix(pyroscope.java): infer service_name label if possible

* Do not mutate target

* update changelog, fix scrape infer

* fix

* fix

CHANGELOG.md

@@ -47,6 +47,8 @@ Main (unreleased)
 - `mimir.rules.kubernetes` has a new `prometheus_http_prefix` argument to configure 
   the HTTP endpoint on which to connect to Mimir's API. (@hainenber)
 
+- `service_name` label is inferred from discovery meta labels in `pyroscope.java` (@korniltsev)
+
 ### Bugfixes
 
 - Fix an issue in `remote.s3` where the exported content of an object would be an empty string if `remote.s3` failed to fully retrieve

component/pyroscope/java/loop.go

@@ -152,6 +152,10 @@ func (p *profilingLoop) push(jfrBytes []byte, startTime time.Time, endTime time.
 		for _, l := range jfrpprofPyroscope.Labels(target, profiles.JFREvent, req.Metric, "", spyName) {
 			ls.Set(l.Name, l.Value)
 		}
+		if ls.Get(labelServiceName) == "" {
+			ls.Set(labelServiceName, inferServiceName(target))
+		}
+
 		profile, err := req.Profile.MarshalVT()
 		if err != nil {
 			_ = l.Log("msg", "failed to marshal profile", "err", err)

component/pyroscope/java/target.go

@@ -0,0 +1,35 @@
+package java
+
+import (
+	"fmt"
+
+	"github.com/grafana/agent/component/discovery"
+)
+
+const (
+	labelServiceName    = "service_name"
+	labelServiceNameK8s = "__meta_kubernetes_pod_annotation_pyroscope_io_service_name"
+)
+
+func inferServiceName(target discovery.Target) string {
+	k8sServiceName := target[labelServiceNameK8s]
+	if k8sServiceName != "" {
+		return k8sServiceName
+	}
+	k8sNamespace := target["__meta_kubernetes_namespace"]
+	k8sContainer := target["__meta_kubernetes_pod_container_name"]
+	if k8sNamespace != "" && k8sContainer != "" {
+		return fmt.Sprintf("java/%s/%s", k8sNamespace, k8sContainer)
+	}
+	dockerContainer := target["__meta_docker_container_name"]
+	if dockerContainer != "" {
+		return dockerContainer
+	}
+	if swarmService := target["__meta_dockerswarm_container_label_service_name"]; swarmService != "" {
+		return swarmService
+	}
+	if swarmService := target["__meta_dockerswarm_service_name"]; swarmService != "" {
+		return swarmService
+	}
+	return "unspecified"
+}

component/pyroscope/scrape/target.go

@@ -430,5 +430,11 @@ func inferServiceName(lset labels.Labels) string {
 	if dockerContainer != "" {
 		return dockerContainer
 	}
+	if swarmService := lset.Get("__meta_dockerswarm_container_label_service_name"); swarmService != "" {
+		return swarmService
+	}
+	if swarmService := lset.Get("__meta_dockerswarm_service_name"); swarmService != "" {
+		return swarmService
+	}
 	return "unspecified"
 }

docs/sources/flow/reference/components/pyroscope.java.md

@@ -62,6 +62,24 @@ The `asprof` binary runs with root permissions.
 If you change the `tmp_dir` configuration to something other than `/tmp`, then you must ensure that the
 directory is only writable by root.
 {{% /admonition %}}
+
+#### `targets` argument
+
+The special `__process_pid__` label _must always_ be present and corresponds to the
+process PID that is used for profiling.
+
+Labels starting with a double underscore (`__`) are treated as _internal_, and are removed prior to scraping.
+
+The special label `service_name` is required and must always be present.
+If it is not specified, `pyroscope.scrape` will attempt to infer it from
+either of the following sources, in this order:
+1. `__meta_kubernetes_pod_annotation_pyroscope_io_service_name` which is a `pyroscope.io/service_name` pod annotation.
+2. `__meta_kubernetes_namespace` and `__meta_kubernetes_pod_container_name`
+3. `__meta_docker_container_name`
+4. `__meta_dockerswarm_container_label_service_name` or `__meta_dockerswarm_service_name`
+
+If `service_name` is not specified and could not be inferred, then it is set to `unspecified`.
+
 ## Blocks
 
 The following blocks are supported inside the definition of

docs/sources/flow/reference/components/pyroscope.scrape.md

@@ -114,6 +114,7 @@ either of the following sources, in this order:
 1. `__meta_kubernetes_pod_annotation_pyroscope_io_service_name` which is a `pyroscope.io/service_name` pod annotation.
 2. `__meta_kubernetes_namespace` and `__meta_kubernetes_pod_container_name`
 3. `__meta_docker_container_name`
+4. `__meta_dockerswarm_container_label_service_name` or `__meta_dockerswarm_service_name`
 
 If `service_name` is not specified and could not be inferred, then it is set to `unspecified`.