[exporter/elasticsearch] Add retry.retry_on_status config (#32585)

Previously, the status codes that trigger retries were hardcoded to be
429, 500, 502, 503, 504.
It is now configurable using `retry.retry_on_status`, and defaults to
`[429, 500, 502, 503, 504]` to avoid a breaking change.
To avoid duplicates, it is recommended to configure
`retry.retry_on_status` to `[429]`, which would be the default in a
future version.

Part of #32584

---------

Co-authored-by: Vishal Raj <[email protected]>
Co-authored-by: Alex Boten <[email protected]>

Author: 3 people

.chloggen/elasticsearchexporter_retry-on-status.yaml

@@ -0,0 +1,30 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: enhancement
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: elasticsearchexporter
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Add retry.retry_on_status config
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [32584]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext: |
+  Previously, the status codes that trigger retries were hardcoded to be 429, 500, 502, 503, 504.
+  It is now configurable using `retry.retry_on_status`, and defaults to `[429, 500, 502, 503, 504]` to avoid a breaking change.
+  To avoid duplicates, it is recommended to configure `retry.retry_on_status` to `[429]`, which would be the default in a future version.
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: [user]

exporter/elasticsearchexporter/README.md

@@ -58,6 +58,7 @@ This exporter supports sending OpenTelemetry logs and traces to [Elasticsearch](
   - `max_requests` (default=3): Number of HTTP request retries.
   - `initial_interval` (default=100ms): Initial waiting time if a HTTP request failed.
   - `max_interval` (default=1m): Max waiting time if a HTTP request failed.
+  - `retry_on_status` (default=[429, 500, 502, 503, 504]): Status codes that trigger request or document level retries. Request level retry and document level retry status codes are shared and cannot be configured separately. To avoid duplicates, it is recommended to set it to `[429]`. WARNING: The default will be changed to `[429]` in the future.
 - `mapping`: Events are encoded to JSON. The `mapping` allows users to
   configure additional mapping rules.
   - `mode` (default=none): The fields naming mode. valid modes are:

exporter/elasticsearchexporter/config.go

@@ -151,6 +151,9 @@ type RetrySettings struct {
 
 	// MaxInterval configures the max waiting time if consecutive requests failed.
 	MaxInterval time.Duration `mapstructure:"max_interval"`
+
+	// RetryOnStatus configures the status codes that trigger request or document level retries.
+	RetryOnStatus []int `mapstructure:"retry_on_status"`
 }
 
 type MappingsSettings struct {

exporter/elasticsearchexporter/config_test.go

@@ -4,6 +4,7 @@
 package elasticsearchexporter
 
 import (
+	"net/http"
 	"path/filepath"
 	"testing"
 	"time"
@@ -61,6 +62,13 @@ func TestLoad_DeprecatedIndexConfigOption(t *testing.T) {
 			MaxRequests:     5,
 			InitialInterval: 100 * time.Millisecond,
 			MaxInterval:     1 * time.Minute,
+			RetryOnStatus: []int{
+				http.StatusTooManyRequests,
+				http.StatusInternalServerError,
+				http.StatusBadGateway,
+				http.StatusServiceUnavailable,
+				http.StatusGatewayTimeout,
+			},
 		},
 		Mapping: MappingsSettings{
 			Mode:  "none",
@@ -136,6 +144,7 @@ func TestLoadConfig(t *testing.T) {
 					MaxRequests:     5,
 					InitialInterval: 100 * time.Millisecond,
 					MaxInterval:     1 * time.Minute,
+					RetryOnStatus:   []int{http.StatusTooManyRequests, http.StatusInternalServerError},
 				},
 				Mapping: MappingsSettings{
 					Mode:  "none",
@@ -186,6 +195,7 @@ func TestLoadConfig(t *testing.T) {
 					MaxRequests:     5,
 					InitialInterval: 100 * time.Millisecond,
 					MaxInterval:     1 * time.Minute,
+					RetryOnStatus:   []int{http.StatusTooManyRequests, http.StatusInternalServerError},
 				},
 				Mapping: MappingsSettings{
 					Mode:  "none",

exporter/elasticsearchexporter/elasticsearch_bulk.go

@@ -103,7 +103,7 @@ func newElasticsearchClient(logger *zap.Logger, config *Config) (*esClientCurren
 		Header:    headers,
 
 		// configure retry behavior
-		RetryOnStatus:        retryOnStatus,
+		RetryOnStatus:        config.Retry.RetryOnStatus,
 		DisableRetry:         retryDisabled,
 		EnableRetryOnTimeout: config.Retry.Enabled,
 		//RetryOnError:  retryOnError, // should be used from esclient version 8 onwards
@@ -175,7 +175,7 @@ func createElasticsearchBackoffFunc(config *RetrySettings) func(int) time.Durati
 	}
 }
 
-func shouldRetryEvent(status int) bool {
+func shouldRetryEvent(status int, retryOnStatus []int) bool {
 	for _, retryable := range retryOnStatus {
 		if status == retryable {
 			return true
@@ -184,15 +184,15 @@ func shouldRetryEvent(status int) bool {
 	return false
 }
 
-func pushDocuments(ctx context.Context, logger *zap.Logger, index string, document []byte, bulkIndexer esBulkIndexerCurrent, maxAttempts int) error {
+func pushDocuments(ctx context.Context, logger *zap.Logger, index string, document []byte, bulkIndexer esBulkIndexerCurrent, maxAttempts int, retryOnStatus []int) error {
 	attempts := 1
 	body := bytes.NewReader(document)
 	item := esBulkIndexerItem{Action: createAction, Index: index, Body: body}
 	// Setup error handler. The handler handles the per item response status based on the
 	// selective ACKing in the bulk response.
 	item.OnFailure = func(ctx context.Context, item esBulkIndexerItem, resp esBulkIndexerResponseItem, err error) {
 		switch {
-		case attempts < maxAttempts && shouldRetryEvent(resp.Status):
+		case attempts < maxAttempts && shouldRetryEvent(resp.Status, retryOnStatus):
 			logger.Debug("Retrying to index",
 				zap.String("name", index),
 				zap.Int("attempt", attempts),

exporter/elasticsearchexporter/factory.go

@@ -8,6 +8,7 @@ package elasticsearchexporter // import "github.com/open-telemetry/opentelemetry
 import (
 	"context"
 	"fmt"
+	"net/http"
 	"runtime"
 	"time"
 
@@ -51,6 +52,13 @@ func createDefaultConfig() component.Config {
 			MaxRequests:     3,
 			InitialInterval: 100 * time.Millisecond,
 			MaxInterval:     1 * time.Minute,
+			RetryOnStatus: []int{
+				http.StatusTooManyRequests,
+				http.StatusInternalServerError,
+				http.StatusBadGateway,
+				http.StatusServiceUnavailable,
+				http.StatusGatewayTimeout,
+			},
 		},
 		Mapping: MappingsSettings{
 			Mode:  "none",

exporter/elasticsearchexporter/integrationtest/datareceiver.go

@@ -137,7 +137,7 @@ func newMockESReceiver(params receiver.CreateSettings, cfg *config, next consume
 
 				if err := next.ConsumeLogs(context.Background(), logs); err != nil {
 					response.HasErrors = true
-					item.Status = http.StatusInternalServerError
+					item.Status = http.StatusTooManyRequests
 					item.Error.Type = "simulated_es_error"
 					item.Error.Reason = err.Error()
 				}

exporter/elasticsearchexporter/logs_exporter.go

@@ -23,14 +23,13 @@ type elasticsearchLogsExporter struct {
 	logstashFormat LogstashFormatSettings
 	dynamicIndex   bool
 	maxAttempts    int
+	retryOnStatus  []int
 
 	client      *esClientCurrent
 	bulkIndexer esBulkIndexerCurrent
 	model       mappingModel
 }
 
-var retryOnStatus = []int{500, 502, 503, 504, 429}
-
 const createAction = "create"
 
 func newLogsExporter(logger *zap.Logger, cfg *Config) (*elasticsearchLogsExporter, error) {
@@ -71,6 +70,7 @@ func newLogsExporter(logger *zap.Logger, cfg *Config) (*elasticsearchLogsExporte
 		index:          indexStr,
 		dynamicIndex:   cfg.LogsDynamicIndex.Enabled,
 		maxAttempts:    maxAttempts,
+		retryOnStatus:  cfg.Retry.RetryOnStatus,
 		model:          model,
 		logstashFormat: cfg.LogstashFormat,
 	}
@@ -129,5 +129,5 @@ func (e *elasticsearchLogsExporter) pushLogRecord(ctx context.Context, resource
 	if err != nil {
 		return fmt.Errorf("Failed to encode log event: %w", err)
 	}
-	return pushDocuments(ctx, e.logger, fIndex, document, e.bulkIndexer, e.maxAttempts)
+	return pushDocuments(ctx, e.logger, fIndex, document, e.bulkIndexer, e.maxAttempts, e.retryOnStatus)
 }

exporter/elasticsearchexporter/logs_exporter_test.go

@@ -328,7 +328,7 @@ func TestExporter_PushEvent(t *testing.T) {
 		server := newESTestServer(t, func(docs []itemRequest) ([]itemResponse, error) {
 			if failures == 0 {
 				failures++
-				return nil, &httpTestError{message: "oops"}
+				return nil, &httpTestError{status: http.StatusTooManyRequests, message: "oops"}
 			}
 
 			rec.Record(docs)
@@ -510,7 +510,7 @@ func withTestExporterConfig(fns ...func(*Config)) func(string) *Config {
 }
 
 func mustSend(t *testing.T, exporter *elasticsearchLogsExporter, contents string) {
-	err := pushDocuments(context.TODO(), zap.L(), exporter.index, []byte(contents), exporter.bulkIndexer, exporter.maxAttempts)
+	err := pushDocuments(context.TODO(), zap.L(), exporter.index, []byte(contents), exporter.bulkIndexer, exporter.maxAttempts, exporter.retryOnStatus)
 	require.NoError(t, err)
 }
 

exporter/elasticsearchexporter/testdata/config.yaml

@@ -19,6 +19,9 @@ elasticsearch/trace:
     bytes: 10485760
   retry:
     max_requests: 5
+    retry_on_status:
+      - 429
+      - 500
 elasticsearch/log:
   tls:
     insecure: false
@@ -38,6 +41,9 @@ elasticsearch/log:
     bytes: 10485760
   retry:
     max_requests: 5
+    retry_on_status:
+      - 429
+      - 500
   sending_queue:
     enabled: true
 elasticsearch/logstash_format: