From 075be0f9578fa4a8e47aebbd3ddbdb1e0494c9db Mon Sep 17 00:00:00 2001
From: mattdurham <mattdurham@ppog.org>
Date: Wed, 27 Dec 2023 14:36:42 -0500
Subject: [PATCH] add max_cache_size (#6026)

* add max_cache_size

* add max_cache_size to converter

* fix tests
---
 CHANGELOG.md                                  |  2 ++
 component/prometheus/relabel/relabel.go       | 20 +++++++++++++------
 component/prometheus/relabel/relabel_test.go  | 14 ++++++++++++-
 .../prometheusconvert/component/relabel.go    |  1 +
 .../components/prometheus.relabel.md          |  3 ++-
 5 files changed, 32 insertions(+), 8 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index f2652fad76ac..260d8437159a 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -74,6 +74,8 @@ Main (unreleased)
 
 - Add support for Basic Auth for agent's web UI, including the APIs. (@hainenber)
 
+- Add `max_cache_size` to `prometheus.relabel` to allow configurability instead of hard coded 100,000. (@mattdurham)
+
 ### 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)
diff --git a/component/prometheus/relabel/relabel.go b/component/prometheus/relabel/relabel.go
index c7b39ad6b0f4..0f3b0c6dc9ac 100644
--- a/component/prometheus/relabel/relabel.go
+++ b/component/prometheus/relabel/relabel.go
@@ -46,15 +46,23 @@ type Arguments struct {
 	MetricRelabelConfigs []*flow_relabel.Config `river:"rule,block,optional"`
 
 	// Cache size to use for LRU cache.
-	//CacheSize int `river:"cache_size,attr,optional"`
+	CacheSize int `river:"max_cache_size,attr,optional"`
 }
 
 // SetToDefault implements river.Defaulter.
-/*func (arg *Arguments) SetToDefault() {
+func (arg *Arguments) SetToDefault() {
 	*arg = Arguments{
-		CacheSize: 500_000,
+		CacheSize: 100_000,
 	}
-}*/
+}
+
+// Validate implements river.Validator.
+func (arg *Arguments) Validate() error {
+	if arg.CacheSize <= 0 {
+		return fmt.Errorf("max_cache_size must be greater than 0 and is %d", arg.CacheSize)
+	}
+	return nil
+}
 
 // Exports holds values which are exported by the prometheus.relabel component.
 type Exports struct {
@@ -88,7 +96,7 @@ var (
 
 // New creates a new prometheus.relabel component.
 func New(o component.Options, args Arguments) (*Component, error) {
-	cache, err := lru.New[uint64, *labelAndID](100_000)
+	cache, err := lru.New[uint64, *labelAndID](args.CacheSize)
 	if err != nil {
 		return nil, err
 	}
@@ -210,7 +218,7 @@ func (c *Component) Update(args component.Arguments) error {
 	defer c.mut.Unlock()
 
 	newArgs := args.(Arguments)
-	c.clearCache(100_000)
+	c.clearCache(newArgs.CacheSize)
 	c.mrc = flow_relabel.ComponentToPromRelabelConfigs(newArgs.MetricRelabelConfigs)
 	c.fanout.UpdateChildren(newArgs.ForwardTo)
 
diff --git a/component/prometheus/relabel/relabel_test.go b/component/prometheus/relabel/relabel_test.go
index c6d2b2699031..d029498555b8 100644
--- a/component/prometheus/relabel/relabel_test.go
+++ b/component/prometheus/relabel/relabel_test.go
@@ -44,11 +44,22 @@ func TestUpdateReset(t *testing.T) {
 	relabeller.relabel(0, lbls)
 	require.True(t, relabeller.cache.Len() == 1)
 	_ = relabeller.Update(Arguments{
+		CacheSize:            100000,
 		MetricRelabelConfigs: []*flow_relabel.Config{},
 	})
 	require.True(t, relabeller.cache.Len() == 0)
 }
 
+func TestValidator(t *testing.T) {
+	args := Arguments{CacheSize: 0}
+	err := args.Validate()
+	require.Error(t, err)
+
+	args.CacheSize = 1
+	err = args.Validate()
+	require.NoError(t, err)
+}
+
 func TestNil(t *testing.T) {
 	ls := labelstore.New(nil, prom.DefaultRegisterer)
 	fanout := prometheus.NewInterceptor(nil, ls, prometheus.WithAppendHook(func(ref storage.SeriesRef, _ labels.Labels, _ int64, _ float64, _ storage.Appender) (storage.SeriesRef, error) {
@@ -72,6 +83,7 @@ func TestNil(t *testing.T) {
 				Action:       "drop",
 			},
 		},
+		CacheSize: 100000,
 	})
 	require.NotNil(t, relabeller)
 	require.NoError(t, err)
@@ -129,7 +141,6 @@ func BenchmarkCache(b *testing.B) {
 
 	lbls := labels.FromStrings("__address__", "localhost")
 	app := entry.Appender(context.Background())
-
 	for i := 0; i < b.N; i++ {
 		app.Append(0, lbls, time.Now().UnixMilli(), 0)
 	}
@@ -161,6 +172,7 @@ func generateRelabel(t *testing.T) *Component {
 				Action:       "replace",
 			},
 		},
+		CacheSize: 100_000,
 	})
 	require.NotNil(t, relabeller)
 	require.NoError(t, err)
diff --git a/converter/internal/prometheusconvert/component/relabel.go b/converter/internal/prometheusconvert/component/relabel.go
index de6ec420ec21..a3bee3c6b7dd 100644
--- a/converter/internal/prometheusconvert/component/relabel.go
+++ b/converter/internal/prometheusconvert/component/relabel.go
@@ -36,6 +36,7 @@ func toRelabelArguments(relabelConfigs []*prom_relabel.Config, forwardTo []stora
 	return &relabel.Arguments{
 		ForwardTo:            forwardTo,
 		MetricRelabelConfigs: ToFlowRelabelConfigs(relabelConfigs),
+		CacheSize:            100_000,
 	}
 }
 
diff --git a/docs/sources/flow/reference/components/prometheus.relabel.md b/docs/sources/flow/reference/components/prometheus.relabel.md
index 93645aec83b6..65cb02394d4a 100644
--- a/docs/sources/flow/reference/components/prometheus.relabel.md
+++ b/docs/sources/flow/reference/components/prometheus.relabel.md
@@ -56,6 +56,7 @@ The following arguments are supported:
 Name | Type | Description | Default | Required
 ---- | ---- | ----------- | ------- | --------
 `forward_to` | `list(MetricsReceiver)` | Where the metrics should be forwarded to, after relabeling takes place. | | yes
+`max_cache_size` | `int` | The maximum number of elements to hold in the relabeling cache. | 100,000 | no
 
 ## Blocks
 
@@ -187,4 +188,4 @@ connection work correctly. Refer to the linked documentation for more details.
 
 {{% /admonition %}}
 
-<!-- END GENERATED COMPATIBLE COMPONENTS -->
\ No newline at end of file
+<!-- END GENERATED COMPATIBLE COMPONENTS -->