Merge branch 'main' into docs/update-topics-concepts

.github/workflows/helm-release.yml

@@ -22,7 +22,7 @@ jobs:
           path: source
 
       - name: Install chart-testing
-        uses: helm/[email protected].0
+        uses: helm/[email protected].1
 
       - name: List changed charts
         id: list-changed

.github/workflows/helm-test.yml

@@ -57,7 +57,7 @@ jobs:
           check-latest: true
 
       - name: Install chart-testing
-        uses: helm/[email protected].0
+        uses: helm/[email protected].1
 
       - name: Determine changed charts
         id: list-changed

CHANGELOG.md

@@ -32,15 +32,33 @@ Main (unreleased)
 
 - `pyroscope.ebpf` support python on arm64 platforms. (@korniltsev)
 
+- `mimir.rules.kubernetes` may now retry its startup on failure. (@hainenber)
+
 - Added links between compatible components in the documentation to make it
   easier to discover them. (@thampiotr)
+
+- Allow defining `HTTPClientConfig` for `discovery.ec2`. (@cmbrad)
+
+- The `remote.http` component can optionally define a request body. (@tpaschalis)
 
 ### Bugfixes
 
 - Update `pyroscope.ebpf` to fix a logical bug causing to profile to many kthreads instead of regular processes https://github.com/grafana/pyroscope/pull/2778 (@korniltsev)
 
 - Update `pyroscope.ebpf` to produce more optimal pprof profiles for python processes https://github.com/grafana/pyroscope/pull/2788 (@korniltsev)
 
+- In Static mode's `traces` subsystem, `spanmetrics` used to be generated prior to load balancing.
+  This could lead to inaccurate metrics. This issue only affects Agents using both `spanmetrics` and 
+  `load_balancing`, when running in a load balanced cluster with more than one Agent instance. (@ptodev)
+
+- Fixes `loki.source.docker` a behavior that synced an incomplete list of targets to the tailer manager. (@FerdinandvHagen)
+
+- Fixes `otelcol.connector.servicegraph` store ttl default value from 2ms to 2s. (@rlankfo)
+
+### Other changes
+
+- Bump github.com/IBM/sarama from v1.41.2 to v1.42.1
+
 v0.38.1 (2023-11-30)
 --------------------
 

cmd/grafana-agent-operator/Dockerfile

@@ -22,15 +22,14 @@ RUN --mount=type=cache,target=/root/.cache/go-build \
     RELEASE_BUILD=${RELEASE_BUILD} VERSION=${VERSION} \
     make operator
 
-FROM ubuntu:lunar
+FROM ubuntu:mantic
 
 LABEL org.opencontainers.image.source="https://github.com/grafana/agent"
 
 # Install dependencies needed at runtime.
 RUN <<EOF
   apt-get update
   apt-get install -qy libsystemd-dev tzdata ca-certificates
-  apt-get upgrade -y
   rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
 EOF
 

cmd/grafana-agent/Dockerfile

@@ -30,15 +30,14 @@ RUN --mount=type=cache,target=/root/.cache/go-build \
     GOEXPERIMENT=${GOEXPERIMENT} \
     make agent
 
-FROM ubuntu:lunar
+FROM ubuntu:mantic
 
 LABEL org.opencontainers.image.source="https://github.com/grafana/agent"
 
 # Install dependencies needed at runtime.
 RUN <<EOF
   apt-get update
   apt-get install -qy libsystemd-dev tzdata ca-certificates
-  apt-get upgrade -y
   rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
 EOF
 

cmd/grafana-agentctl/Dockerfile

@@ -23,15 +23,14 @@ RUN --mount=type=cache,target=/root/.cache/go-build \
     GO_TAGS="netgo promtail_journal_enabled" \
     make agentctl
 
-FROM ubuntu:lunar
+FROM ubuntu:mantic
 
 LABEL org.opencontainers.image.source="https://github.com/grafana/agent"
 
 # Install dependencies needed at runtime.
 RUN <<EOF
   apt-get update
   apt-get install -qy libsystemd-dev tzdata ca-certificates
-  apt-get upgrade -y
   rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
 EOF
 

component/discovery/aws/ec2.go

@@ -7,6 +7,7 @@ import (
 	"github.com/aws/aws-sdk-go/aws/ec2metadata"
 	"github.com/aws/aws-sdk-go/aws/session"
 	"github.com/grafana/agent/component"
+	"github.com/grafana/agent/component/common/config"
 	"github.com/grafana/agent/component/discovery"
 	"github.com/grafana/river/rivertypes"
 	promcfg "github.com/prometheus/common/config"
@@ -42,18 +43,21 @@ type EC2Arguments struct {
 	RefreshInterval time.Duration     `river:"refresh_interval,attr,optional"`
 	Port            int               `river:"port,attr,optional"`
 	Filters         []*EC2Filter      `river:"filter,block,optional"`
+
+	HTTPClientConfig config.HTTPClientConfig `river:",squash"`
 }
 
 func (args EC2Arguments) Convert() *promaws.EC2SDConfig {
 	cfg := &promaws.EC2SDConfig{
-		Endpoint:        args.Endpoint,
-		Region:          args.Region,
-		AccessKey:       args.AccessKey,
-		SecretKey:       promcfg.Secret(args.SecretKey),
-		Profile:         args.Profile,
-		RoleARN:         args.RoleARN,
-		RefreshInterval: model.Duration(args.RefreshInterval),
-		Port:            args.Port,
+		Endpoint:         args.Endpoint,
+		Region:           args.Region,
+		AccessKey:        args.AccessKey,
+		SecretKey:        promcfg.Secret(args.SecretKey),
+		Profile:          args.Profile,
+		RoleARN:          args.RoleARN,
+		RefreshInterval:  model.Duration(args.RefreshInterval),
+		Port:             args.Port,
+		HTTPClientConfig: *args.HTTPClientConfig.Convert(),
 	}
 	for _, f := range args.Filters {
 		cfg.Filters = append(cfg.Filters, &promaws.EC2Filter{

component/discovery/aws/ec2_test.go

@@ -0,0 +1,29 @@
+package aws
+
+import (
+	"net/url"
+	"testing"
+
+	"github.com/grafana/agent/component/common/config"
+	"github.com/stretchr/testify/require"
+	"gotest.tools/assert"
+)
+
+func TestConvert(t *testing.T) {
+	// parse example proxy
+	u, err := url.Parse("http://example:8080")
+	require.NoError(t, err)
+	httpClientConfig := config.DefaultHTTPClientConfig
+	httpClientConfig.ProxyURL = config.URL{URL: u}
+
+	// example configuration
+	riverArgs := EC2Arguments{
+		Region:           "us-east-1",
+		HTTPClientConfig: httpClientConfig,
+	}
+
+	// ensure values are set
+	promArgs := riverArgs.Convert()
+	assert.Equal(t, "us-east-1", promArgs.Region)
+	assert.Equal(t, "http://example:8080", promArgs.HTTPClientConfig.ProxyURL.String())
+}

component/faro/receiver/sourcemaps.go

@@ -18,7 +18,7 @@ import (
 	"github.com/go-sourcemap/sourcemap"
 	"github.com/grafana/agent/component/faro/receiver/internal/payload"
 	"github.com/grafana/agent/pkg/flow/logging/level"
-	"github.com/minio/pkg/wildcard"
+	"github.com/grafana/agent/pkg/util/wildcard"
 	"github.com/prometheus/client_golang/prometheus"
 	"github.com/vincent-petithory/dataurl"
 )

component/loki/source/docker/docker.go

@@ -242,11 +242,11 @@ func (c *Component) Update(args component.Arguments) error {
 			return err
 		}
 		targets = append(targets, tgt)
-
-		// This will never fail because it only fails if the context gets canceled.
-		_ = c.manager.syncTargets(context.Background(), targets)
 	}
 
+	// This will never fail because it only fails if the context gets canceled.
+	_ = c.manager.syncTargets(context.Background(), targets)
+
 	c.args = newArgs
 	return nil
 }

component/loki/source/windowsevent/bookmark.go

@@ -8,11 +8,12 @@ package windowsevent
 import (
 	"bytes"
 	"errors"
-	"github.com/natefinch/atomic"
 	"io"
 	"io/fs"
 	"os"
 
+	"github.com/natefinch/atomic"
+
 	"github.com/grafana/loki/clients/pkg/promtail/targets/windows/win_eventlog"
 )
 
@@ -55,6 +56,7 @@ func newBookMark(path string) (*bookMark, error) {
 	if err != nil {
 		return nil, err
 	}
+	defer file.Close()
 	fileContent, err := io.ReadAll(file)
 	if err != nil {
 		return nil, err

component/mimir/rules/kubernetes/rules.go

@@ -10,6 +10,7 @@ import (
 	"github.com/grafana/agent/component"
 	"github.com/grafana/agent/pkg/flow/logging/level"
 	mimirClient "github.com/grafana/agent/pkg/mimir/client"
+	"github.com/grafana/dskit/backoff"
 	"github.com/grafana/dskit/instrument"
 	promListers "github.com/prometheus-operator/prometheus-operator/pkg/client/listers/monitoring/v1"
 	"github.com/prometheus/client_golang/prometheus"
@@ -153,10 +154,22 @@ func New(o component.Options, args Arguments) (*Component, error) {
 }
 
 func (c *Component) Run(ctx context.Context) error {
-	err := c.startup(ctx)
-	if err != nil {
-		level.Error(c.log).Log("msg", "starting up component failed", "err", err)
-		c.reportUnhealthy(err)
+	startupBackoff := backoff.New(
+		ctx,
+		backoff.Config{
+			MinBackoff: 1 * time.Second,
+			MaxBackoff: 10 * time.Second,
+			MaxRetries: 0, // infinite retries
+		},
+	)
+	for {
+		if err := c.startup(ctx); err != nil {
+			level.Error(c.log).Log("msg", "starting up component failed", "err", err)
+			c.reportUnhealthy(err)
+		} else {
+			break
+		}
+		startupBackoff.Wait()
 	}
 
 	for {
@@ -205,8 +218,7 @@ func (c *Component) startup(ctx context.Context) error {
 	if err := c.startRuleInformer(); err != nil {
 		return err
 	}
-	err := c.syncMimir(ctx)
-	if err != nil {
+	if err := c.syncMimir(ctx); err != nil {
 		return err
 	}
 	go c.eventLoop(ctx)

component/otelcol/connector/servicegraph/servicegraph.go

@@ -91,7 +91,7 @@ var DefaultArguments = Arguments{
 	Dimensions: []string{},
 	Store: StoreConfig{
 		MaxItems: 1000,
-		TTL:      2 * time.Millisecond,
+		TTL:      2 * time.Second,
 	},
 	CacheLoop:           1 * time.Minute,
 	StoreExpirationLoop: 2 * time.Second,

component/otelcol/connector/servicegraph/servicegraph_test.go

@@ -44,7 +44,7 @@ func TestArguments_UnmarshalRiver(t *testing.T) {
 				Dimensions: []string{},
 				Store: servicegraphprocessor.StoreConfig{
 					MaxItems: 1000,
-					TTL:      2 * time.Millisecond,
+					TTL:      2 * time.Second,
 				},
 				CacheLoop:           1 * time.Minute,
 				StoreExpirationLoop: 2 * time.Second,

component/remote/http/http.go

@@ -41,6 +41,7 @@ type Arguments struct {
 
 	Method  string            `river:"method,attr,optional"`
 	Headers map[string]string `river:"headers,attr,optional"`
+	Body    string            `river:"body,attr,optional"`
 
 	Client common_config.HTTPClientConfig `river:"client,block,optional"`
 }
@@ -193,7 +194,12 @@ func (c *Component) pollError() error {
 	ctx, cancel := context.WithTimeout(context.Background(), c.args.PollTimeout)
 	defer cancel()
 
-	req, err := http.NewRequest(c.args.Method, c.args.URL, nil)
+	var body io.Reader
+	if c.args.Body != "" {
+		body = strings.NewReader(c.args.Body)
+	}
+
+	req, err := http.NewRequest(c.args.Method, c.args.URL, body)
 	if err != nil {
 		level.Error(c.log).Log("msg", "failed to build request", "err", err)
 		return fmt.Errorf("building request: %w", err)

component/remote/http/http_test.go

@@ -3,6 +3,7 @@ package http_test
 import (
 	"context"
 	"fmt"
+	"io"
 	"net/http"
 	"net/http/httptest"
 	"sync"
@@ -27,6 +28,9 @@ func Test(t *testing.T) {
 	defer srv.Close()
 
 	handler.SetHandler(func(w http.ResponseWriter, r *http.Request) {
+		b, err := io.ReadAll(r.Body)
+		require.NoError(t, err)
+		require.Equal(t, string(b), "hello there!")
 		fmt.Fprintln(w, "Hello, world!")
 	})
 
@@ -40,10 +44,11 @@ func Test(t *testing.T) {
             "x-custom" = "value",
 			"User-Agent" = "custom_useragent",
         }
+		body = "%s"
 
 		poll_frequency = "50ms" 
 		poll_timeout   = "25ms" 
-	`, srv.URL, http.MethodPut)
+	`, srv.URL, http.MethodPut, "hello there!")
 	var args http_component.Arguments
 	require.NoError(t, river.Unmarshal([]byte(cfg), &args))
 

docs/docs_updated_test.go

@@ -1,3 +1,5 @@
+//go:build !windows
+
 package docs
 
 import (

docs/sources/flow/monitoring/component_metrics.md

@@ -13,30 +13,21 @@ weight: 200
 
 # Component metrics
 
-{{< param "PRODUCT_NAME" >}} [components][] may optionally expose Prometheus metrics
-which can be used to investigate the behavior of that component. These
-component-specific metrics are only generated when an instance of that
-component is running.
-
-> Component-specific metrics are different than any metrics being processed by
-> the component. Component-specific metrics are used to expose the state of a
-> component for observability, alerting, and debugging.
-
-Component-specific metrics are exposed at the `/metrics` HTTP endpoint of the
-{{< param "PRODUCT_NAME" >}} HTTP server, which defaults to listening on
-`http://localhost:12345`.
-
-> The documentation for the [`grafana-agent run`][grafana-agent run] command describes how to
-> modify the address {{< param "PRODUCT_NAME" >}} listens on for HTTP traffic.
-
-Component-specific metrics will have a `component_id` label matching the
-component ID generating those metrics. For example, component-specific metrics
-for a `prometheus.remote_write` component labeled `production` will have a
-`component_id` label with the value `prometheus.remote_write.production`.
-
-The [reference documentation][] for each component will describe the list of
-component-specific metrics that component exposes. Not all components will
-expose metrics.
+{{< param "PRODUCT_NAME" >}} [components][] may optionally expose Prometheus metrics which can be used to investigate the behavior of that component.
+These component-specific metrics are only generated when an instance of that component is running.
+
+> Component-specific metrics are different than any metrics being processed by the component.
+> Component-specific metrics are used to expose the state of a component for observability, alerting, and debugging.
+
+Component-specific metrics are exposed at the `/metrics` HTTP endpoint of the {{< param "PRODUCT_NAME" >}} HTTP server, which defaults to listening on `http://localhost:12345`.
+
+> The documentation for the [`grafana-agent run`][grafana-agent run] command describes how to > modify the address {{< param "PRODUCT_NAME" >}} listens on for HTTP traffic.
+
+Component-specific metrics have a `component_id` label matching the component ID generating those metrics.
+For example, component-specific metrics for a `prometheus.remote_write` component labeled `production` will have a `component_id` label with the value `prometheus.remote_write.production`.
+
+The [reference documentation][] for each component described the list of component-specific metrics that the component exposes.
+Not all components expose metrics.
 
 {{% docs/reference %}}
 [components]: "/docs/agent/ -> /docs/agent/<AGENT_VERSION>/flow/concepts/components.md"