The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
honeycomb
exporter: Update to use internal data format (#1689)
newrelic
exporter: Add support for span events (#1643)awsemf
exporter:- Add placeholder support in
log_group_name
andlog_stream_name
config (#1623, #1661) - Add label matching filtering rule (#1619)
- Add placeholder support in
resourcedetection
processor: Add new resource detector for AWS Elastic Beanstalk environments (#1585)loadbalancing
exporter:- Add sort of endpoints in static resolver (#1692)
- Allow specifying port when using DNS resolver (#1650)
- Add
batchperresourceattr
helper library that splits an incoming data based on an attribute in the resource (#1694) alibabacloudlogservice
exporter:- Add logs exporter (#1609)
- Change trace type from opencensus to opentelemetry (#1713)
datadog
exporter:- Improve trace exporter performance (#1706, #1707)
- Add option to only send metadata (#1723)
awsxray
exporter:- Add parsing of Python stack traces (#1676)
- Add collector version to user agent (#1730)
loadbalancing
exporter:- Fix retry queue for exporters (#1687)
- Fix
periodicallyResolve
for DNS resolver checks (#1678)
datadog
exporter: Fix status code handling (#1691)awsxray
exporter:- Fix empty traces in X-Ray console (#1709)
- Stricter requirements for adding http request url (#1729)
- Fix status code handling for errors/faults (#1740)
signalfx
exporter:- Split incoming data requests by access token before enqueuing (#1727)
- Disable retry on 400 and 401, retry with backoff on 429 and 503 (#1672)
awsecscontainermetrics
receiver: Improve error handling to fix seg fault (#1738)
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
zookeeper
receiver: Collects metrics from a Zookeeper instance using themntr
commandloadbalacing
exporter: Consistently exports spans belonging to the same trace to the same backendwindowsperfcounters
receiver: Captures the configured system, application, or custom performance counter data from the Windows registry using the PDH interfaceawsprometheusremotewrite
exporter: Sends metrics data in Prometheus TimeSeries format to a Prometheus Remote Write Backend and signs each outgoing HTTP request following the AWS Signature Version 4 signing process
awsemf
exporter:- Add
metric_declarations
config option for metric filtering and dimensions (#1503) - Add SummaryDataType and remove Min/Max from Histogram (#1584)
- Add
signalfxcorrelation
exporter: Add ability to translate host dimension (#1561)newrelic
exporter: Use pdata instead of the OpenCensus for traces (#1587)metricstransform
processor:- Add
combine
action for matched metrics (#1506) - Add
submatch_case
config option to specify case of matched label values (#1640)
- Add
awsecscontainermetrics
receiver: Extract cluster name from ARN (#1626)elastic
exporter: Improve handling of span status if the status code is unset (#1591)
awsemf
exporter: Add check for unhandled metric data types (#1493)groupbytrace
processor: Make buffered channel to avoid goroutines leak (#1505)stackdriver
exporter: Setoptions.UserAgent
so that the OpenCensus exporter does not override the UA ($1620)
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
datadog
exporter to send metric and trace data to Datadog (#1352)tailsampling
processor moved from core to contrib (#1383)
jmxmetricsextension
migrated tojmxreceiver
(#1182, #1357)- Move signalfx correlation code out of
sapm
tosignalfxcorrelation
exporter (#1376) - Move Splunk specific utils outside of common (#1306)
stackdriver
exporter:- Config options
metric_prefix
&skip_create_metric_descriptor
are now nested undermetric
, see README. - Trace status codes no longer reflect gRPC codes as per spec changes: open-telemetry/opentelemetry-specification#1067
- Config options
datadog
exporter: Remove option to change the namespace prefix (#1483)
splunkhec
receiver: Add ability to ingest metrics (#1276)signalfx
receiver: Improve pipeline error handling (#1329)datadog
exporter:- Improve hostname resolution (#1285)
- Add flushing/export of traces and trace-related statistics (#1266)
- Enable traces on Windows (#1340)
- Send otel.exporter running metric (#1354)
- Add tag normalization util method (#1373)
- Send host metadata (#1351)
- Support resource conventions for hostnames (#1434)
- Add version tag extract (#1449)
- Add
batchpertrace
library to split the incoming batch into several batches, one per trace (#1257) statsd
receiver:- Add timer support (#1335)
- Add sample rate support for counter, transfer gauge to double and transfer counter to int only (#1361)
awsemf
exporter: Restructure metric translator logic (#1353)resourcedetection
processor:- Add EC2 hostname attribute (#1324)
- Add ECS Resource detector (#1360)
sapm
exporter: Add queue settings (#1390)metrictransform
processor: Add metric filter option (#1447)awsxray
exporter: Improve ECS attribute and origin translation (#1428)resourcedetection
processor: Initial system detector (#1405)
- Remove duplicate definition of cloud providers with core conventions (#1288)
kubeletstats
receiver: Handle nil references from the kubelet API (#1326)awsxray
receiver:- Add kind type to root span to fix the empty parentID problem (#1338)
- Fix the race condition issue (#1490)
awsxray
exporter:- Setting the tlsconfig InsecureSkipVerify using NoVerifySSL (#1350)
- Drop invalid xray trace id (#1366)
elastic
exporter: Ensure span name is limited (#1371)splunkhec
exporter: Don't send 'zero' timestamps to Splunk HEC (#1157)stackdriver
exporter: Skip processing empty metrics slice (#1494)
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
sapm
exporter:- Enable queuing by default (#1224)
- Add SignalFx APM correlation (#1205)
- Make span source attribute and destination dimension names configurable (#1286)
signalfx
exporter:- Pass context to the http client requests (#1225)
- Update
disk.summary_utilization
translation rule to accommodate new labels (#1258)
newrelic
exporter: Addspan.kind
attribute (#1263)datadog
exporter:- Add Datadog trace translation helpers (#1208)
- Add API key validation (#1216)
splunkhec
receiver: Add the ability to ingest logs (#1268)awscontainermetrics
receiver: ReportCpuUtilized
metric in percentage (#1283)awsemf
exporter: Only calculate metric rate for cumulative counter and avoid SingleDimensionRollup for metrics with only one dimension (#1280)
- Make
signalfx
exporter a metadata exporter (#1252) awsecscontainermetrics
receiver: Check for empty network rate stats and set zero (#1260)awsemf
exporter: Remove InstrumentationLibrary dimension in CloudWatch EMF Logs if it is undefined (#1256)awsxray
receiver: Fix trace/span id transfer (#1264)datadog
exporter: Remove trace support for Windows for now (#1274)sapm
exporter: Correlation enabled check inversed (#1278)
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
awsemf
exporter to support exporting metrics to AWS CloudWatch (#498, #1169)http_forwarder
extension that forwards HTTP requests to a specified target (#979, #1014, #1150)datadog
exporter that sends metric and trace data to Datadog (#1142, #1178, #1181, #1212)awsecscontainermetrics
receiver to collect metrics from Amazon ECS Task Metadata Endpoint (#1089, #1148, #1160)
signalfx
exporter:- Add host metadata synchronization (#1039, #1118)
- Add
copy_dimensions
translator option (#1126) - Update
k8s_cluster
metric translations (#1121) - Add option to exclude metrics (#1156)
- Add
avg
aggregation method (#1151) - Fallback to host if cloud resource id not found (#1170)
- Add backwards compatible translation rules for the
dockerstatsreceiver
(#1201) - Enable queuing and retries (#1223)
splunkhec
exporter:- Add log support (#875)
- Enable queuing and retries (#1222)
k8scluster
receiver: Standardize metric names (#1119)awsxray
exporter:- Support AWS EKS attributes (#1090)
- Store resource attributes in X-Ray segments (#1174)
honeycomb
exporter:- Add span kind to the event sent to Honeycomb (#474)
- Add option to adjust the sample rate using an attribute on the span (#1162)
jmxmetrics
extension: Add subprocess manager to manage child java processes (#1028)elastic
exporter: Initial metrics support (#1173)k8s
processor: Rename default attr names for label/annotation extraction (#1214)- Add common SignalFx host id extraction (#1100)
- Allow MSI upgrades (#1165)
awsxray
exporter: Don't set origin to EC2 when not on AWS (#1115)
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
- add
dockerstats
receiver as top level component (#1081) - add
tracegen
utility (#956)
stackdriver
exporter: Allow overriding client options via config (#1010)k8scluster
receiver: Ensure informer caches are synced before initial data sync (#842)elastic
exporter: Translatedeployment.environment
resource attribute to Elastic APM's semantically equivalentservice.environment
(#1022)k8s
processor: Add logs support (#1051)awsxray
exporter: Log response error with zap (#1050)signalfx
exporter- Add dimensions to renamed metrics (#1041)
- Add translation rules for
disk_ops.total
anddisk_ops.pending
metrics (#1082) - Add event support (#1036)
kubeletstats
receiver: Cache detailed PVC labels to reduce API calls (#1052)signalfx
receiver: Add event support (#1035)
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
- add initial docker stats receiver, without sourcing in top level components (#495)
- add initial jmx metrics extension structure, without sourcing in top level components (#740)
routing
processor for routing spans based on HTTP headers (#907)splunkhec
receiver to receive Splunk HEC metrics, traces and logs (#840)- Add skeleton for
http_forwarder
extension that forwards HTTP requests to a specified target (#979)
stackdriver
exporter- Add timeout parameter (#835)
- Add option to configurably set UserAgent string (#758)
signalfx
exporter- Reduce memory allocations for big batches processing (#871)
- Add AWSUniqueId and gcp_id generation (#829)
- Calculate cpu.utilization compatibility metric (#839, #974, #954)
metricstransform
processor: Replace{{version}}
in label values (#876)resourcedetection
processor: Logs Support (#970)statsd
receiver: Add parsing for labels and gauges (#903)
k8s
processor- Wrap metrics before sending further down the pipeline (#837)
- Fix setting attributes on metrics passed from agent (#836)
awsxray
exporter: Fix "pointer to empty string" is not omitted bug (#830)azuremonitor
exporter: Treat UNSPECIFIED span kind as INTERNAL (#844)signalfx
exporter: Remove misleading warnings (#869)newrelic
exporter: Fix panic if service name is empty (#969)honeycomb
exporter: Don't emit default proc id + starttime (#972)
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
- Remove deprecated
lightstep
exporter (#828)
statsd
receiver for ingesting StatsD messages (#566)
signalfx
exporter- Add disk usage translations (#760)
- Add disk utilization translations (#782)
- Add translation rule to drop redundant metrics (#809)
kubeletstats
receiver- Sync available volume metadata from /pods endpoint (#690)
- Add ability to collect detailed data from PVC (#743)
awsxray
exporter: Translate SDK name/version into xray model (#755)elastic
exporter: Translate semantic conventions to Elastic destination fields (#671)stackdriver
exporter: Add point count metric (#757)awsxray
receiver- Ported the TCP proxy from the X-Ray daemon (#774)
- Convert to OTEL trace format (#691)
kubeletstats
receiver: Do not break down metrics batch (#754)host
observer: Fix issue on darwin where ports listening on all interfaces are not correctly accounted for (#582)newrelic
exporter: Fix panic on missing span status (#775)
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
- Receivers
prometheusexec
subprocess manager (##499)
signalfx
exporter- Add/Update metric translations (#579, #584, #639, #640, #652, #662)
- Add support for calculate new metric translator (#644)
- Add renaming rules for load metrics (#664)
- Update
container.name
tok8s.container.name
in default translation rule (#683) - Rename working-set and page-fault metrics (#679)
awsxray
exporter- Translate exception event into xray exception (#577)
- Add ingestion of X-Ray segments via UDP (#502)
- Parse Java stacktrace and populate in xray cause (#687)
kubeletstats
receiver- Add metric_groups option (#648)
- Set datapoint timestamp in receiver (#661)
- Change
container.name
label tok8s.container.name
(#680) - Add working-set and page-fault metrics (#666)
- Add basic support for volume metrics (#667)
stackdriver
trace exporter: Move to new interface and pdata (#486)metricstranform
processor: Keep timeseries and points in order after aggregation (#663)k8scluster
receiver: Changecontainer.spec.name
label tok8s.container.name
(#681)- Migrate receiver creator to internal data model (#701)
- Add ec2 support to
resourcedetection
processor (#587) - Enable timeout, sending queue and retry for SAPM exporter (#707)
azuremonitor
exporter: Correct HTTP status code success mapping (#588)k8scluster
receiver: Fix owner reference in metadata updates (#649)awsxray
exporter: Fix handling of db system (#697)
- Skeleton for AWS ECS container metrics receiver (#463)
prometheus_exec
receiver (#655)
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
awsxray
receiver updated to support udp:tcp_endpoint
config option renamed toendpoint
(#497)- TLS config changed for
sapmreceiver
(#488) andsignalfxreceiver
receivers (#488)
- Exporters
sentry
adds tracing exporter for Sentry (#565)
- Extensions
endpoints
observer: adds generic endpoint watcher (#427)host
observer: looks for listening network endpoints on host (#432)
- Update
honeycomb
exporter for v0.8.0 compatibility - Extend
metricstransform
processor to be able to add a label to an existing metric (#441) - Update
kubeletstats
metrics according to semantic conventions (#475) - Updated
awsxray
receiver config to use udp (#497) - Add
/pods
endpoint support inkubeletstats
receiver to add extra labels (#569) - Add metric translation options to
signalfx
exporter (#477, #501, #571, #573)
azuremonitor
exporter: Mark spanToEnvelope errors as permanent (#500)
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
- Removed
jaegarlegacy
(#397) andzipkinscribe
receivers (#410) kubeletstats
receiver: Renamedk8s.pod.namespace
pod label tok8s.namespace.name
andk8s.container.name
container label tocontainer.name
- Processors
metricstransform
renames/aggregates within individual metrics (#376) and allow changing the data type between int and float (#402)
awsxray
exporter: Usepeer.service
as segment name when set. (#385)splunk
exporter: Add trace exports support (#359, #399)- Build and publish Windows MSI (#408) and DEB/RPM Linux packages (#405)
kubeletstats
receiver:- Fixed NPE for newly created pods (#404)
- Updated to latest change in the ReceiverFactoryOld interface (#401)
- Fixed logging and self reported metrics (#357)
awsxray
exporter: Only convert SQL information for SQL databases. (#379)resourcedetection
processor: Correctly obtain machine-type info from gce metadata (#395)k8scluster
receiver: Fix container resource metrics (#416)
Released 01-07-2020
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
- Processors
resourcedetection
to automatically detect the resource based on the configured set of detectors (#309)
kubeletstats
receiver: Support for ServiceAccount authentication (#324)signalfx
exporter and receiver- Add SignalFx metric token passthrough and config option (#325)
- Set default endpoint of
signalfx
receiver to:9943
(#351)
awsxray
exporter: Support aws plugins EC2/ECS/Beanstalk (#343)sapm
exporter and receiver: Add SAPM access token passthrough and config option (#349)k8s
processor: Add metrics support (#358)k8s
observer: Separate annotations from labels in discovered pods (#363)
honeycomb
exporter: Remove shared use of libhoney from goroutines (#305)
Released 17-06-2020
The OpenTelemetry Collector Contrib contains everything in the opentelemetry-collector release (be sure to check the release notes here as well!). Check out the Getting Started Guide for deployment and configuration information.
signalfx
exporterurl
parameter changed toingest_url
(no impact if only usingrealm
setting)
- Receivers
receiver_creator
to create receivers at runtime (#145), add observer support to receiver_creator (#173), add rules support (#207), add dynamic configuration values (#235)kubeletstats
receiver (#237)prometheus_simple
receiver (#184)kubernetes-cluster
receiver (#175)redis
receiver (#138)
- Exporters
alibabacloudlogservice
exporter (#259)SplunkHEC
metrics exporter (#246)elastic
APM exporter (#240)newrelic
exporter (#229)
- Extensions
k8s
observer (#185)
awsxray
exporter- Use X-Ray convention of segment name == service name (#282)
- Tweak xray export to improve rendering of traces and improve parity (#241)
- Add handling for spans received with nil attributes (#212)
honeycomb
exporter- Use SendPresampled (#291)
- Add span attributes as honeycomb event fields (#271)
- Support resource labels in Honeycomb exporter (#20)
k8s
processor- Add support of Pod UID extraction to k8sprocessor (#219)
- Use
k8s.pod.ip
to record resource IP instead of justip
(#183) - Support same authentication mechanism as other kubernetes components do (#307)
sapm
exporter: Add TLS for SAPM and SignalFx receiver (#215)signalfx
exporter- Add metric metadata syncer to SignalFx exporter (#231)
- Add TLS for SAPM and SignalFx receiver (#215)
stackdriver
exporter: Add support for resource mapping in config (#163)
awsxray
exporter: Wrap bad request errors for proper handling by retry queue (#205)lightstep
exporter: Ensure Lightstep exporter doesnt crash on nil node (#250)sapm
exporter: Do not break Jaeger traces before sending downstream (#193)k8s
processor: Ensure Jaeger spans work in passthrough mode (262)
Traces | Metrics |
---|---|
Jaeger Legacy | Carbon |
SAPM (SignalFx APM) | Collectd |
Zipkin Scribe | K8s Cluster |
Redis | |
SignalFx | |
Simple Prometheus | |
Wavefront |
- K8s
Commercial | Community |
---|---|
Alibaba Cloud Log Service | Carbon |
AWS X-ray | Elastic |
Azure Monitor | Jaeger Thrift |
Honeycomb | Kinesis |
Lightstep | |
New Relic | |
SAPM (SignalFx APM) | |
SignalFx (Metrics) | |
Splunk HEC | |
Stackdriver (Google) |
- Observer
- K8s
Released 2020-03-30
- Make prometheus receiver config loading strict. #697 Prometheus receiver will now fail fast if the config contains unused keys in it.
- Enable best effort serve by default of Prometheus Exporter (orijtech/prometheus-go-metrics-exporter#6)
- Fix null pointer exception in the logging exporter #743
- Remove unnecessary condition to have at least one processor #744
- Updated Honeycomb exported to
honeycombio/opentelemetry-exporter-go v0.3.1
Receivers / Exporters:
- AWS X-Ray
- Carbon
- CollectD
- Honeycomb
- Jaeger
- Kinesis
- LightStep
- OpenCensus
- OpenTelemetry
- SAPM
- SignalFx
- Stackdriver
- Wavefront
- Zipkin
- Zipkin Scribe
Processors:
- Attributes
- Batch
- Memory Limiter
- Queued Retry
- Resource
- Sampling
- Span
- Kubernetes
Extensions:
- Health Check
- Performance Profiler
- zPages
Released 2020-03-25
Alpha v0.2.8 of OpenTelemetry Collector Contrib.
- Implemented OTLP receiver and exporter.
- Added ability to pass config to the service programmatically (useful for custom builds).
- Improved own metrics / observability.
Released 2020-03-17
- New command-line switch to control legacy and new metrics. Users are encouraged to experiment and migrate to the new metrics.
- Improved error handling on shutdown.
- Fixed passthrough mode k8sprocessor.
- Added
HASH
action to attribute processor.
- Added Honeycomb exporter.
- Added LightStep exporter.
- Added regular expression for Carbon receiver, allowing the metric name to be broken into proper label keys and values.
- Updated Stackdriver exporter to use a new batch API.
Released 2020-02-18
- Updated metrics prefix to
otelcol
and expose command line argument to modify the prefix value. - Batch dropped span now emits zero when no spans are dropped.
- Extended Span processor to have include/exclude span logic.
- Ability to choose strict or regexp matching for include/exclude filters.
- Added Carbon receiver and exporter.
- Added Wavefront receiver.
Released 2020-01-30
- Regexp-based filtering of span names.
- Ability to extract attributes from span names and rename span.
- File exporter for debugging.
- Span processor is now enabled by default.
Released 2020-01-11
First release of OpenTelemetry Collector Contrib.