Skip to content

Latest commit

 

History

History
3079 lines (2439 loc) · 182 KB

CHANGELOG.md

File metadata and controls

3079 lines (2439 loc) · 182 KB
Raw

Changelog

v0.63.0

🛑 Breaking changes 🛑

  • processor/cumulativetodelta: Sets the processor.cumulativetodeltaprocessor.EnableHistogramSupport feature gate to enabled by default. Histograms will be converted by default if they match include rules. (#15288)

  • pkg/ottl: Renames accessor for instrumentation scope from instrumentation_library to instrumentation_scope to match documentation (#14933)

  • pkg/ottl: Add lists to the OTTL grammar and change the function signature of Concat. (#13391) The following functions have changed: keep_keys now has a function signature of keep_keys(target, keys[]). Concat now has a function signature of Concat(keys[], delimiter).

  • pkg/ottl: Updates ParseStatements to return a []*Statement instead of []Statement. (#14911)

  • postgresqlreceiver: Change postgresql.bgwriter.duration data type from int to double (#14725, #14972)

  • extension/headerssetter: Remove deprecated extension/headerssetter module. (#13774) The headerssetter extension was superseded by headerssetterextension.

  • elasticsearchreceiver: removing direction feature gate (#14955)

  • hostmetricsreceiver: remove direction feature gate (#14959)

  • kubeletstatsreceiver: remove direction feature gate (#14961)

  • memcachedreceiver: removing direction feature gate (#14964)

  • vcenterreceiver: removing direction feature gate (#14963)

  • zookeeperreceiver: removing direction feature gate (#14962)

  • exporter/routingprocessor: Rename OTTL expression configuration parameter of the routing table to statement to align with the OTTL naming. (#14950)

  • dockerstatsreceiver: container.cpu.usage.system is no longer a default metric. Added more description for the metric. (#9794, #14558) Change is not breaking unless you are have enabled the featuregate receiver.dockerstats.useScraperV2 and are using the specified metric.

  • processor/transform: Convert the keep_keys and Concat functions to use list parameters and change the function signature of Concat. (#13391) The updated functions now have the following signatures: keep_keys now has a function signature of keep_keys(target, keys[]). Concat now has a function signature of Concat(keys[], delimiter).

🚩 Deprecations 🚩

  • exporter/influxdb: change status to beta (#14098)
  • receiver/influxdb: change status to beta (#14099)

🚀 New components 🚀

  • azureblobreciver: Add a new component azureblobreciver (#8834) Add a new component azureblobreceiver that reads logs and traces from Azure Blob Storage
  • k8sobjectsreceiver: Add a new k8sobjects receiver to collect(pull/watch) Kubernetes Objects (#14185)
  • httpcheckreceiver: New HTTP Check receiver allows users to run synthethic checks from the collector (#10607)

💡 Enhancements 💡

  • oracledbreceiver: Add metrics captured by this receiver (#13939)
  • hostmetricsreceiver: Add a new metric process.paging.faults to the process scraper of the hostmetrics receiver. (#14084)
  • receiver/prometheusreceiver: Append exemplars to the metrics received by prometheus receiver (#8353) Acknowledge exemplars coming from prometheus receiver and append it to otel format
  • azureblobreceiver: Implementation of the component (second PR) (#8834)
  • azureeventhubreceiver: Adds implementation of Azure event hub receiver for raw data (#12786)
  • bearertokenauthextension: Allow bearertokenauthextension to support custom auth schemes apart from "Bearer" (#14771)
  • clickhouseexporter: Add support export OTLP traces to ClickHouse (#8028)
  • mysqlreceiver: add mysql.opened_resources metric (#14138)
  • receiver/mysql: add metrics based on events_statements_summary_by_digest table (#14138)
  • mysqlreceiver: add mysql.mysqlx_worker_threads (#14138)
  • receiver/statsdreceiver: Add OTLP exponential histogram aggregator support for high-resolution histogram and timing metrics (#5742)
  • pkg/translator/loki: Add support for formatting Loki log lines in logfmt format, toggled by providing loki.format hint. It can have value logfmt or json. JSON is default if no format hint is present. (#15351)
  • exporter/lokiexporter: Add multi-tenancy support for Loki exporter via loki.tenant hint. (#14706)
  • pkg/translator/loki: Add support for grouping Loki requests by attribute that is resolved from the loki.tenant hint. (#14706)
  • cmd/mdatagen: Add ability to see if enabled option is overridden in user settings. (#15344)
  • oracledbreceiver: Adds DML locks and transaction metrics to capture usage and limits (#13939)
  • pkg/ottl: Add new Instrumentation Scope context to allow for efficient transformation of Instrumentation Scope telemetry. (#14892)
  • pkg/ottl: Add new Metric context to allow for efficient transformation of metric telemetry. (#14895)
  • snmpreceiver: adds SNMP config helper for SNMP Receiver metric scraper (#13409)
  • snmpreceiver: adds otel metric helper struct for SNMP metric receiver scraper (#13409)
  • snmpreceiver: changes the client of the SNMP metric receiver to only return data in its functions rather than try to process that data (#13409)
  • snmpreceiver: changes the config of the SNMP Metric Receiver to have "double" as an option in place of "float" (#13409)
  • solacereceiver: Updates Solace Receiver with a variety of improvements and fixes for compatibility with Solace PubSub+ Event Broker 10.2.0 (#15244)
  • syslogreceiver: Added RFC 6587 Octet Counting and Non-Transparent-Framing support. (#8390)
  • elasticsearchexporter: upgrade version of elasticsearch client to 8.4 (#15385)
  • dockerstatsreceiver: Log warning when using ScraperV1, and add documentation for migrating from V1 to V2. (#9794, #14968)
  • windowsperfcountersreceiver: Retry counter retrieval once, when error indicates possible rollover (#14343)

🧰 Bug fixes 🧰

  • servicegraph: Add servicegraph processor to component list (#14899) Bumps servicegraph stability to alpha
  • azuremonitorexporter: Application Insights Severity Level mapping was incorrectly based on SeverityText which was unique for every language/SDK. (#15275) Application Insights Severity Level mapping was incorrectly based on SeverityText which was unique for every language/SDK. The mapping approach has been changed to analyze SeverityNumber, which is supposed to be common across languages/SDKs.
  • awscloudwatchreceiver: Fix config validation for some named configurations. (#14952)
  • exporter/awsxrayexporter: Fix an issue where the awsxrayexporter would not parse .Net stacktrace correctly. (#14780)
  • postgresqlreceiver: Fix issue where WAL lag stats could cause panic (#14972)
  • prometheusremotewriteexporter: Fix value of +Inf histogram bucket to be equal to total count (#4975)
  • prometheusreceiver: Fix metrics being grouped into the same metrics family incorrectly (#13117)
  • spanmetricsprocessor: Sets TraceID and SpanID fields in Exemplar type (as per the spec) and removes the use of FilteredAttributes to pass these values around. (#13401)
  • splunkhecexporter: Skip SummaryDataPoint.Sum and HistogramDataPoint.Sum NaN values (#14877)
  • pkg/winperfcounters: Fix testing with -race flag for winperfcounters package and winperfcounter based receivers (#10145, #10146, #10149, #10150)

v0.62.0

🛑 Breaking changes 🛑

  • pkg/ottl: Enhance replace_all_patterns functions to take a new parameter that specifies whether the function applies to keys or values. (#12631)
  • pkg/ottl: Unexports several types intended only for the package's grammar. (#14599)
  • pkg/ottl: Changed Statement to be an interface with an Execute function. (#14869)
  • pkg/ottl: Rename the Queries signal configuration field to Statements and remove the ottlconfig package. (#14680)
  • confmap/provider/s3provider: Rename s3mapprovider to a s3provider making it consistent with other providers. (#14616)
  • processor/transform: Rename the queries config key to statements. (#14680)

🚀 New components 🚀

  • oracledbreceiver: Add a new Oracle DB receiver component collecting metrics by periodically querying the database. (#13939)
  • azuredataexplorer: Adding a new exporter to OpenTelemetry that supports Metrics, Logs and Traces into Azure Data Explorer, a timeseries & analytics offering from Azure (#11830)
  • awscloudwatchreceiver: Adds the awscloudwatchreceiver in an alpha state (#14449)
  • snmpreceiver: adds the snmpreceiver metric receiver (#13409)

💡 Enhancements 💡

  • probabilisticsamplerprocessor: add more mterics to probabilistic sampler for observ. (#13924)
  • mongodbatlasreceiver: Safeguard against large amounts of alert emissions. (#14610)
  • bearertokenauthextension: support reading tokens from file (#14325)
  • cumulativetodeltaprocessor: Reduce memory consumption in cumulativetodeltaprocessor by removing unnecessary storage of metric identities. (#13751)
  • carbonexporter: Change carbonexporter to use pdata, remove dependency on opencensus (#14589)
  • coralogixexporter: Adds support for mapping application name and subsystem using Resource attributes (#14398)
  • mysqlreceiver: add mysql.tmp_resources metric (#14138)
  • mysqlreceiver: add mysql.locked_connects metric (#14138)
  • elasticsearchreceiver: Add scraping index stats and generate search metrics (#14635)
  • pkg/telemetryquerylanguage: Add the Int factory function to the transform processor (#11810)
  • exporters: Use BuildInfo.Command for identifying the collector in some AWS exporter user agents. (#14719) Some exporters were using a build-time constant definition to change the identity of the collector binary in user agent strings. These will now use the collector service's BuildInfo.Command value.
  • jaegerexporter: Adds support for ValueTypeBytes to Jaeger's trace to jaegerproto translator. (#14574)
  • mongodbatlasreceiver: Adds capability to retrieve alerts from the API rather than listening (#14121)
  • pkg/ottl: Use generics to avoid context cast in getters and funcs. (#14482)
  • pkg/ottl: Simplify parser creation logic. (#14601)
  • pkg/ottl: Add new Resource context to allow for efficient transformation of resource telemetry. (#14887)
  • exporter/signalfxexporter/translation: Drop datapoints that have more than 36 dimensions and log a message when agent is set to debug (#14625) Additionally, the metric and dimension validation failures are now logged in Debug instead of Warn
  • sigv4authextension: Have "sts_region" default to "region" if not provided (#14573)
  • snmpreceiver: adds the client for the snmpreceiver (#13409)
  • snmpreceiver: adds the full configuration that the SNMP Receiver will use (#13409)
  • splunkhecreceiver: Updates Splunk receiver http status codes in order to be compliant with SplunkCloud (#14469)
  • pkg/stanza: readerFactory and Reader use helper.Encoding directly, no longer depends on helper.Splitter (#14593)
  • pkg/stanza: readerFactory and Reader use bufio.SplitFunc directly, no longer depends on helper.Splitter (#14766)
  • pkg/stanza: add splitter factory which return a split func (#14766)
  • exporter/awsxrayexporter: Change the value of xraysegment.url from dbConnectionString to the span name. This makes the XRay segment timeline more informationally useful. (#14342) This change contravenes the AWS documentation for what values should go into this segment field.
  • pkg/translator/zipkin: Change zipkin V1 conversion to use pdata. (#14592)
  • filterexpr: Prevent the matcher from panicking (#13573)

🧰 Bug fixes 🧰

  • clickhouseexporter: Fix serviceName variable scope (#8028)
  • exporter/datadogexporter: Fix an issue where the Datadog system metrics system.memory.{usage,total} would be incorrect in Kubernetes. (#14618)
  • mysqlreceiver: Fix sql queries for scraping table and index metrics. (#14138)
  • fileexporter: set rotation disabled by default (#14690)
  • fluentforwardreceiver: added case to handle uint64 timestamp (#11435)
  • sqlqueryreceiver: fix oracle db integration test (#12332)
  • prometheusreceiver: changes to use the new scrape_configs endpoint in the target allocator to dynamically pull scrape configuration. (#14597)
  • saphanareceiver: Fix incorrect use of units for saphana licenses metrics (#10565)
  • hostmetricsreceiver: Allow to continue collecting partition information even if an error is returned (#14315) Log the error and continue if partition information, however incomplete, was provided.
  • kubeletstatsreceiver: Fix use network io as network errors bug (#14318)

v0.61.0

🛑 Breaking changes 🛑

  • dockerstatsreceiver: For V2 scraper implementation, change the defaults from emitting everything to a sensible set of defaults. (#9794, #14093) This is only breaking if you have explicitly enabled the featuregate receiver.dockerstats.useScraperV2.
  • dockerstatsreceiver: Change 'stats' config key to 'metrics'. (#9794, #14184) Note: this is only breaking for those who have explicitly enabled the featuregate receiver.dockerstatsd.useScraperV2.
  • iisreceiver: Emit metrics per-site and per-app-pool with new resource attributes. (#14448)
  • pkg/telemetryquerylanguage: Unexport BoolExpressionEvaluator, Literal, NewGetter, and NewFunctionCall. (#13737)
  • mongodbatlasreceiver: Retain actual raw log line as Body. The raw attribute is now removed. Use Body instead for the raw log line. (#14178)
  • mysqlreceiver: The metrics are now being emitted with a resource attribute marking the endpoint of the database. (#14138)
  • pkg/ottl: Rename ParseQueries to ParseStatements and rename Query to Statement. (#14444)
  • jaegerreceiver: Remove remote sampling endpoint from the Jaeger receiver (#6633)
  • oteltransformationlanguage: Renames ottldatapoints to ottldatapoints to better represent how the context is intended to be used. (#14384)
  • pkg/ottl: Renames Telemetry Query Language to OpenTelemetry Transformation Language to avoid confusion with an analysis language. (#14150)
  • pkg/ottl: Flattens ottlcommon and ottlotel into ottlfuncs. (#14386)
  • pkg/ottl: Replaced Logger with TelemetrySettings (#14389)

🚩 Deprecations 🚩

  • coralogixexporter: deprecating jaeger based tracing configuration in favour of OpenTelemetry protocol based one. (#7931) Please remove the endpoint field and use the new traces.endpoint field with your OpenTelemetry endpoint.
  • elasticsearchreceiver, hostmetricsreceiver, kubelestatsreceiver, memcachedreceiver, vcenterreceiver, zookeeperreceiver: Log message to announce direction attribute feature gate deprecation (#14129) The change to remove the direction attribute has been reverted in the specification. As a result, the following feature gates will be removed in v0.62.0:
    • receiver.elasticsearchreceiver.emitMetricsWithDirectionAttribute
    • receiver.elasticsearchreceiver.emitMetricsWithoutDirectionAttribute
    • receiver.hostmetricsreceiver.emitMetricsWithDirectionAttribute
    • receiver.hostmetricsreceiver.emitMetricsWithoutDirectionAttribute
    • receiver.kubelestatsreceiver.emitMetricsWithDirectionAttribute
    • receiver.kubelestatsreceiver.emitMetricsWithoutDirectionAttribute
    • receiver.memcachedreceiver.emitMetricsWithDirectionAttribute
    • receiver.memcachedreceiver.emitMetricsWithoutDirectionAttribute
    • receiver.vcenterreceiver.emitMetricsWithDirectionAttribute
    • receiver.vcenterreceiver.emitMetricsWithoutDirectionAttribute
    • receiver.zookeeperreceiver.emitMetricsWithDirectionAttribute
    • receiver.zookeeperreceiver.emitMetricsWithoutDirectionAttribute
  • extensions/headerssetter: Deprecate the extensions/headerssetter module, use extensions/headerssetterextenions instead. (#13774)

🚀 New components 🚀

  • s3mapprovider: A new implementation of ConfigMapProvider for Amazon S3 (s3mapprovider) allows OTEL Collector the ability to load configuration for itself by fetching and reading config files stored in Amazon S3. (#12939)

💡 Enhancements 💡

  • loki: Split the conversion OTLP -> Loki into its own package (#13649)
  • kafkareceiver: Add direct unmarshaler, inserting Kafka payload bytes as the body of a log record (#13252)
  • apachereceiver: The receiver now supports 6 more metrics, more information in the linked issue. (#14095)
  • datadogexporter: Add logs support (#2651)
  • datadogexporter: The "hostmetrics" receiver metrics are now correctly mapped to Datadog system metrics. (#14005)
  • fileexporter: support for compressing the telemetry data before exporting. (#13626)
  • fileexporter: support for encoding telemetry data using proto marshaler (#13626)
  • googlecloudspannerreceiver: Fixed errors when transaction-stats columns are NULL (#14189)
  • cumulativetodeltaprocessor: Reduce memory consumption of histograms in cumulativetodeltaprocessor by allocating only a single identity per datapoint. (#13751)
  • chronyreceiver: Improved chronyreceiver's context and timeout handling (#14131) Updated chronyreceiver to use net.Dialer.DialContext and respect timeouts more closely
  • filterspan: Add span kind filtering. (#13612)
  • mysqlreceiver: The receiver now scraper 16 new metrics related to io_waits. (#14138)
  • filelog, journald, syslog, tcplog, udplog, windowseventlog: Allow 'parse_to' fields to accept 'attributes' and 'resource' (#14089)
  • sentryexporter: Make sentry status more detailed - use HTTP and Grpc codes from tags (#13407)
  • routingprocessor: Add support for using OpenTelemetry Transformation Language (OTTL) expressions as routing conditions. (#13158)
  • transformprocessor: Add Split function, which separates a string by the delimiter and returns an array of substrings. (#11790)
  • pkg/winperfcounters: Add counter path to error if scraping fails (#14443)

🧰 Bug fixes 🧰

  • awsemfexporter: Fix regression that causes null dimensions to be appended during dimension roll up. (#14532)
  • k8sprocessor: check every association for eventual update (#13119)
  • sentryexporter: Omit empty parent span id (#13415)
  • pkg/ottl: Replace invalid comparison print statement with a debug log entry (#14467)
  • servicegraphprocessor: Fixed metric names to match the specification. (#14187)
  • hostmetricsreceiver: On Windows systems, do not fail to start if a performance counter is missing or inaccessible. (#14032)
  • loki: Fix converting log resources to loki labels for more than one log record. (#14288)
  • mongodbatlasreceiver: Fix timestamp parsing for mongodb 4.2 audit logs (#14168)
  • mongodbatlasreceiver: Add missing fields for audit logs (#14177)
  • mongodbatlasreceiver: Fix potential infinite loop when scraping audit logs (#14169)
  • otlpjsonfilereceiver: logs receive operations should use StartLogsOp/EndLogsOp (#14535)
  • pkg/translator/prometheus: do not normalize metric name with colon (#14135)

v0.60.0

🛑 Breaking changes 🛑

  • pkg/stanza: Change name of ParserConfig.Config to more specific ParserConfig.SeverityConfig (#14126)

  • pkg/stanza/adapter: Remove OperatorConfigs type, rename LogReceiverType.DecodeInputConfig to LogReceiverType.InputConfig. (#14078)

  • postgresqlreceiver: The receiver will now emit resource attributes via default. (#13811) old behavior can be established by setting the featuregates receiver.postgresql.emitMetricsWithResourceAttributes to false and receiver.postgresql.emitMetricsWithoutResourceAttributes to true

  • k8sattributesprocessor: Change the way how key_regex setting is handled. After this change, provided expressions are applied to the full length of attribute values. (#9716)

  • pkg/telemetryquerylanguage: Add ability to specify attribute keys in the limit function that aren't allowed to be dropped (#9734) This breaking change affects the transform processor since we don't have a way to default values in a function call.

🚩 Deprecations 🚩

  • exporter/influxdb: Change status to unmaintained (#14098)
  • receiver/influxdb: Change status to unmaintained (#14099)

🚀 New components 🚀

  • instanaexporter: Enable Instana exporter (#13395)
  • servicegraph: Add traces processor that builds a map representing the interrelationships between various services in a system (#9232)

💡 Enhancements 💡

  • elasticsearchreceiver: Add additional node metric for cache count (#14027)

  • pkg/telemetryquerylanguage: Add Concat, which allows concatenating an arbitrary number of strings with a delimiter (#12476)

  • processor/filter: Added ability to filter by metric type using expression MetricType keyword. (#14107)

  • elasticsearchreceiver: Add additional two cluster-level metrics (#13748) Add additional cluster metrics for elasticsearch.cluster.in_flight_fetch & elasticsearch.cluster.pending_tasks.

  • dockerstatsreceiver: Simplify BlockIO metrics by taking operation out of the metric name and putting it in an attribute. (#9794, #13445)

  • elasticsearchreceiver: Added node.fs.disk.{free,total} to elasticsearch receiver (#13571)

  • elasticsearchexporter: Support sending OpenTelemetry tracing data to Elasticsearch index or data stream. (#12069)

  • elasticsearchreceiver: Adds new method Version in the client request to get the elasticsearch version number (#14012)

  • fileexporter: Support for rotation of telemetry files (#13626)

  • processor/filer: Add support for ExponentialHistogram and Summary metrics (#14116)

  • receiver/k8scluster: Moving initialization of k8s client from receiver init to receiver start phase. (#12582)

  • lokiexporter: Overhaul of the Loki Exporter, in preparation for native OTLP support at Loki (#12873)

  • awskinesisexporter: Upgrading client to use v2 of the AWS SDK (#13898)

  • prometheusreceiver: Remove storing pointers to pdata points, allow removing unnecessary pdata points copy. (#13922)

  • sigv4authextension: Add more context to credential errors generated by the sigv4auth extension. (#14031)

  • tailsampling: Add support for all sampling policies in and & composite policy (#11505)

  • pkg/telemetryquerylanguage: Add a Logger interface and allow it to be specified as an argument type in TQL functions. (#9730)

  • telemetryquerylanguage: Adds inequality operators <, <=, >=, > to TQL. (#12491)

  • telemetryquerylanguage: Add the Int factory function. (#11810)

  • telemetryquerylanguage: Add split factory function to separate a string by the delimiter, and returns an array of substrings. (#11790)

  • processor/transform: Add Concat, which allows concatenating an arbitrary number of strings with a delimiter (#12476)

🧰 Bug fixes 🧰

  • hostmetricsreceiver: Set process metrics start time to process create time (#11447)
  • awsemfexporter: Properly handle empty dimension set in metric_declarations (#13766)
  • clickhouseexporter: cast FlagsStruct into uint32 in ExecContext to fix export failure (#13843)
  • k8sprocessor: fix the passthrough mode (#13765) Treating k8s.pod.ip resource attribute in special way, as it is added by passthrough mode
  • elasticsearchreceiver: Fix issue where elasticsearch.os.* metrics were not being collected (#13983)
  • elasticsearchreceiver: Fix issue where elasticsearch.node.script.* metrics were not being collected (#13983)
  • elasticsearchreceiver: Fix issue where elasticsearch.node.translog.* metrics were not being collected (#13983)
  • elasticsearchreceiver: Fix issue where elasticsearch.cluster_state.* metrics were not being collected (#13930)
  • elasticsearchreceiver: Fix issue where elasticsearch.node.shards.data_set.size emits 0 value for unsupported version (#14012)
  • elasticsearchreceiver: Fix elasticsearch.node.disk.io.read/write metrics from By to KiBy (#13815)
  • elasticsearchreceiver: Fix issue where elasticsearch.indexing_pressure.memory.limit emits 0 value for unsupported version (#14012)
  • elasticsearchreceiver: Fix issue where elasticsearch.cluster.state_update.time emits attributes pairs that don't exist (#13984)
  • awsxrayexporter: Fixes a bug in the logic for parsing CloudWatch Log Group ARNs (#13702)
  • loadbalancingexporter: Fixed the case in loadbalancingexporter where a DNS resolver is defined without an explicit Port. (#10321)
  • k8sattributesprocessor: k8sattributesprocessor now correctly adds attributes to new pods that start after the collector (#13619)
  • tailsamplingprocessor: Fixes SpanCount sampler not using the correct span count for decision. (#13865)
  • filelogreceiver: Fix issue for missing key_value_parser operator (#13631)
  • mongodbatlasreceiver: fix log parsing for clusters using major version 4.2 (#14008)
  • chronyreceiver: When trying to read from socket, the socket type was incorrect (#13862)
  • postgresqlreceiver: Uses the postgres databasename when retrieving database inventory information (#13641)
  • prometheusreceiver/prometheusremotewriteexporter: Leave the sum unset in histograms without sums, and don't produce _sum metric points for histograms without sums (#7546)
  • processor/redaction: Update redaction attributes in case if data sent through the processor more than once, not not ignore them. (#13854)
  • chloggen: changelog generation tool moved to https://github.com/open-telemetry/opentelemetry-go-build-tools (#14022)
  • exporter/AlibabaCloudLogServiceExporter: Fix issue that promethus occuring error when the resource metric labels contains dot (#3429)
  • exporter/tanzuobservabilityexporter: This change causes tanzuobservabilityexporter to depend on 0.10.4 of the wavefront-sdk-go library. (#13417)
  • transformprocessor: Fixes panic of transformprocessor handling Gauge types (#13905)
  • vcenterreceiver: Fix incorrect KBy and MBy units, updating them to KiBy and MiBy (#13935)

v0.59.0

🛑 Breaking changes 🛑

  • elasticsearchreceiver: Remove direction for metrics. The feature gate: receiver.elasticsearchreceiver.emitMetricsWithoutDirectionAttribute can be set to apply the following (#13258) (#12189)

    • elasticsearch metrics:
      • elasticsearch.node.cluster.io will become:
        • elasticsearch.node.cluster.io.received
        • elasticsearch.node.cluster.io.sent

  • kubeletstatsreceiver: Remove direction for metrics. The feature gate: receiver.kubeletstatsreceiver.emitMetricsWithoutDirectionAttribute can be set to apply the following (#12588) (#12164)

    • kubeletstats metrics:
      • k8s.node.network.io will become:
        • k8s.node.network.io.transmit
        • k8s.node.network.io.receive
      • k8s.node.network.errors will become:
        • k8s.node.network.errors.transmit
        • k8s.node.network.errors.receive

  • processor/metricstransform: Remove processor.metricstransformprocessor.UseOTLPDataModel feature flag. OpenCensus-based implementation is not available anymore. (#13405)

  • exporter/mezmoexporter: This change removes the hardcoded "otel" hostname that was embedded in outgoing logs data. (#13410) It is replaced by:

    1. Sending to a new collector endpoint that does not require the hostname parameter.

    2. Recognizing the "host.name" resource attribute and using that value to fill in log metadata recognized upstream.

    This is a breaking change, and as such will generate a startup error if the exporter is configured to send to an endpoint that does not support this feature.

  • filelog, journald, otlpjsonfile, syslog, tcplog, udplog, windowseventlog receivers: Remove helper.Duration struct. Configured time values no longer have a default unit of s (#13439)

  • pkg/stanza: Remove HostIdentifier, HostIdentifierConfig and associated functions (#13396)

  • filelog, journald, syslog, tcplog, udplog, and windowseventlog receivers: Fix a bug where original severity text was not preserved. (#13263)

  • pkg/stanza: Update function and struct names in keyvalue parser (#13418)

  • cmd/mdatagen: Remove old metadata generator (#13374)

🚩 Deprecations 🚩

  • jaegerreceiver: Add remote sampling deprecation warning (#6633)

🚀 New components 🚀

  • pulsarreceiver: Add Apache Pulsar receiver (#9796)
  • instanaexporter: Add Instana exporter implementation (#13395)
  • instanaexporter: Add initial structure of the Instana exporter (#13395)

💡 Enhancements 💡

  • loadbalancingexporter: Added interval and timeout to the dns configuration in the loadbalancer exporter (#10199)

  • pkg/telemetryquerylanguage: Add Join, which allows joining an arbitrary number of strings with a delimiter (#12476)

  • elasticsearchreceiver: Add additional metrics (#13115) Add additional metrics for indexing_pressure, published_states, state_queue, ingest, script.

  • hostmetricsreceiver: Add threads count metric (#12482)

  • aerospikereceiver: Add query_count and query_tracked metrics (#13255)

  • docsgen: use contrib components, add makefile target (#12639)

  • dockerstatsreceiver: Change relevant memory metrics from gauges to sums, so they are aligned with spec recommendations and can be aggregated. (#9794)

  • pkg/stanza/csv_parser: Add ignore_quotes option that ignores all quoting in fields if true. (#13656)

  • pkg/telemetryquerylanguage: Add TQL mappings for InstrumentationScope.attributes (#13639)

  • k8sattributesprocessor: Add support for discovering Kubernetes CronJob name (#141)

  • coreinternal: Move test utilities out of testbed into coreinternal, allow use of these lightweight helpers from other modules. (#5742)

  • kafkametricsreceiver: Migrate receiver to the new metrics builder (#7142) This allows users to disable particular metrics in through user settings.

  • mongodbatlasreceiver: Add logs retrieval capability (#12347)

  • mongodbreceiver: Enhance partial error handling with descriptive metric/attribute messages. (#13367)

  • filelog, journald, syslog, tcplog, udplog, windowseventlog receivers: Add ability to set log body when when parsing. (#10274)

  • filelog, journald, syslog, tcplog, udplog, windowseventlog receivers: Enhance error message when csv_parser finds unexpected number of fields (#13427)

  • filelog, journald, syslog, tcplog, udplog, windowseventlog receivers: Enable debugging operators stdout and file_output (#13394)

  • postgresqlreceiver: Adds background writer metrics (#13327) Adds the following metrics:

    • postgresql.bgwriter.buffers.allocated
    • postgresql.bgwriter.buffers.writes
    • postgresql.bgwriter.checkpoint.count
    • postgresql.bgwriter.duration
    • postgresql.bgwriter.maxwritten.count

  • postgresqlreceiver: Adds more metrics regarding database tables. (#13228) metrics added: postgresql.database.count postgresql.table.count postgresql.table.size postgresql.table.vacuum.count

  • postgresqlreceiver: Adds WAL and connections metrics (#13399) metrics added

    • postgresql.replication.data_delay
    • postgresql.wal.age
    • postgresql.wal.lag
    • postgresql.connection.max

  • prometheusreceiver: allow to query scrape jobs from OpenTelemetryOperators TargetAllocator (#7944)

  • prometheusremotewrite: Support customization of target_info metric including the ability to disable it (#12300)

  • prometheusexporter: Adds a feature to prometheusexporter that enables it to export exemplars along with histogram metrics. (#5192)

  • prometheusexporter: Use HTTPServerSettings in the exporter to leverage inbuilt TLS. (#10851)

  • receivercreator: adds the unique EndpointID to generated receiver componentID to prevent collisions (#12670)

  • spanmetricsprocessor: The unit of the latency metric is now explicitly defined. (#13423)

  • extension/storage/storagetest: Add in-memory and file-backed test extensions and clients (#13086)

  • telemetryquerylanguage: Avoid copying attributes when filtering on keys in the keep_keys TQL function (#4756)

  • prometheusreceiver: Remove temporary maps to dedup and remove unuseful labels (#13705)

🧰 Bug fixes 🧰

  • processor/metricstransform: Aggregate cumulative data points with different start time. (#12611)
  • expvarreceiver: Fixes an bug where the mSpanSys value was recorded into the mCacheSys metric. (#13171)
  • extension/filestorage: Allow putting the compaction temp folder on a different filesystem. (#13449)
  • prometheusremotewriteexporter: Handle the case with 0 metrics gracefully. (#10364)
  • prometheusreceiver: Fix num data point for metrics recorded in prometheusreceiver (#13705)
  • loadbalancingexporter: Fixed a crash if endpointFor is called before any endpoints have been discovered (#10110)
  • prometheusreceiver: Fix timestamp for histograms and summaries with no Sum (#13705)
  • exporter/sumologic: Mark the exporter as mutating. (#13647)
  • jaegerremotesampling: Mark as alpha (#13005)
  • mongodbreceiver: Adds metric versioning checks to prevent partial errors (#13155)
  • mongodbreceiver: Change lpu integration test to use a mongodb 4.4 (#12981)
  • pkg/translator/prometheus: do not normalize metric name with multiple underscore (#13322)
  • prometheusreceiver: Make the error returned when dropping summary metrics that collide with histograms clearer. (#12976)
  • prometheusremotewriteexporter: Don't emit the target_info metric when it would only contain job and instance (#12768)
  • receiver_creator: Correct observer instance reference by updating watch_observers use ComponentID instead of just Type (#12801)
  • vcenterreceiver: Re-establish sessions when running with particular TLS settings. (#13447)

v0.58.0

🛑 Breaking changes 🛑

🚩 Deprecations 🚩

  • dotnetdiagnosticsreceiver: Change status to unmaintained (#12757)

🚀 New components 🚀

  • azureblobreceiver: Add a new component azureblobreciver (#8834) Add a new component azureblobreceiver that reads logs and traces from Azure Blob Storage
  • azureeventhubreceiver: New component to receive logs from Azure Event Hubs (#12786)
  • headers_setter: Add support for setting exporter headers from the upstream requests context (#5733, #7945, #4814)

💡 Enhancements 💡

  • filterprocessor: Add ability to filter based on logs SeverityText and Body. (#9235)
  • aerospikereceiver: Support tls configuration. Add tls and tlsname config items. (#12770)
  • datadogexporter: Send host info when sending hostmetadata from datadogexporter (#12944)
  • jaegerremotesampling: extend jaegerremotesampling with remote sampling via gRPC (#6694)
  • filterprocessor: Add log filtering based on SeverityNumber (#12959)
  • hostmetricsreceiver: Add new resource attribute process.parent_pid. The value of the attribute is the parent PID (#12290)
  • k8sattributesprocessor: Add support for discovering Kubernetes DaemonSet name and DaemonSet UID" (#141)
  • k8sattributesprocessor: Add support for discovering Kubernetes Job name and Job UID" (#141)
  • k8sattributesprocessor: Add support for discovering Kubernetes ReplicaSet name and ReplicaSet UID (#141)
  • k8sattributesprocessor: Add support for discovering Kubernetes StatefulSet name and StatefulSet UID" (#141)
  • loadbalancingexporter: Exporting trace pipelines based on Service name to avoid duplicate label svc + operation in collector hosts. (#12652)
  • lokiexporter: Don't retry on 4xx responses (excluding 429) from loki (#12930)
  • observers: Adds multiple observer.Notify subscriber support to observer.EndpointWatcher for multiple Endpoint event consumers. (#10830, #11541, #11544)
  • filelogreceiver: Process batches consecutively, rather than one per poll interval (#10285)
  • postgresqlreceiver: Client refactor to eventually support resource attributes. (#13087)
  • postgresqlreceiver: Adds Index Resource Attribute and Metrics Collection. (#13167) This enhancement will only be enabled with the receiver.postgresql.emitMetricsWithResourceAttributes feature gate being enabled.
  • pkg/translator/signalfx: report Histogram min/max when present, do not report sum when not present (#13153)
  • sigv4authextension: Added separate region for STS for cross region auth (#12736)
  • solacereceiver: Adds topic destination as a new attribute to receiver's spans (#12640)
  • pkg/telemetryquerylanguage: Add Getter slices to the TQL parser, allowing them to be specified as function parameters. (#12476)

🧰 Bug fixes 🧰

  • aerospikereceiver: Fix crash when connecting to an Aerospike server with authentication using a config with collect_cluster_metrics == false and no username or password. (#12979)
  • apachereceiver: Fix some partial errors not being correctly reported (#13133)
  • couchdbreceiver: Fix some partial errors not being correctly reported (#13007)
  • signalfxreceiver: log attempts to divide by zero at the debug level (#12969)
  • mongodbreceiver: Fix partial scraper errors not being reported (#13108)
  • mysqlreceiver: Fix some partial errors not being correctly reported (#13009)
  • tailsamplingprocessor: Fixing null pointer exception for span count sampling strategy. (#12745)

v0.57.2

🛑 Breaking changes 🛑

  • healthcheckextension: Remove deprecated port field from config. (#12668)
  • processor/cumulativetodelta: Removes the deprecated metrics config option. Use include/exclude instead. (#12732)
  • lokiexporter: Deprecate the format option (#12897)
  • metricstransformprocessor: Remove deprecated metric_name settings option (#12737)
  • pkg/stanza/fileconsumer: Unexport several fields that are meant for internal usage only (#12793)
  • pkg/stanza/fileconsumer: Rename fileconsumer.Input to fileconsumer.Manager (#12876)
  • prometheusremotewriteexporter: Export target_info not target for resource attributes (#12079)
  • pkg/translator/signalfx: Remove no longer necessary prom compatible configuration from FromTranslator (#12671)
  • signalfxexporter: Remove no longer necessary feature gate for prom compatible metrics (#12671)
  • zookeeperreceiver: Remove direction for metrics. The feature gate: receiver.zookeeperreceiver.emitMetricsWithoutDirectionAttribute can be set to apply the following (#12772) (#12184)
    • zookeeper metrics:
      • zookeeper.packet.count will become:
        • zookeeper.packet.received.count
        • zookeeper.packet.sent.count

🚀 New components 🚀

  • chronyreceiver: This will add in support for measuring tracking data from chronyd (#11789)

💡 Enhancements 💡

  • datadogexporter: Add host metadata tags to metrics on serverless environments (#12486)

  • observer: Add observer.Endpoint.ID to Env() (#12751)

  • aerospikereceiver: Scrape nodes in parallel when in collect_cluster_metrics mode. (#11563)

  • coralogixexporter: Add logs client to coralogixexporter (#12601)

  • awscontainerinsightreceiver: Pod Detection changes to support Containerd runtime in K8s (#12638)

  • dockerstatsreceiver: Enable a featuregate for the new implementation of scrape. Change the units in the new implementation to align to semantic convention. (#9794) This allows the new implementation of scrape to be used via a featuregate. It also updates the tests, which now test for expected descriptions and units in the new implementation.

  • healthcheckextension: Use confighttp.HTTPServerSettings to allow tls and auth. (#12668)

  • demo: Add connection time out for client in examples/demo/client (#12431)

  • prometheusexporter: Export target_info metrics with the resource attributes. (#8265)

  • resourcedetectionprocessor: update Azure resource detector to save "name" from Azure metadata API in azure.vm.name attribute (#12779)

  • pkg/telemetryquerylanguage: Adds standard contexts to the module that should be used when interacting with OTLP traces, metrics, and logs. (#12589)

  • pkg/telemetryquerylanguage: Adds a set of otel functions to the telemetryquerylanguage module. (#12754)

  • pkg/telemetryquerylanguage: Adds a set of common functions to the telemetryquerylanguage module. (#12739)

  • cumulativetodeltaprocessor: Cumulative Histogram metrics will now be converted to delta temporality and are no longer skipped. (#12423) This means that consumers who currently rely on Histograms being skipped would need to ensure they are excluded via config.

  • dockerstatsreceiver: Initial PR for onboarding dockerstats onto mdatagen scraping. (#9794) Additionally appends the internal docker client to allow options (bug fix).

  • mongodbreceiver: Enhance mongodbreceiver by adding new metrics (#12672)

  • chronyreceiver: Added internal chrony client (#11789) -| Added an internal client for the receiver with modifications of the upstream client to fit the collector

  • receivercreator: add per-receiver resource_attribute and validate endpoint type keys on global (#11766)

  • pkg/translator/signalfx: Optimize ToMetrics to emit datapoints under the same metric when same name and type. (#12902)

  • sqlqueryreceiver: Add static_attributes optional configuration to allow adding attributes/tags to queried metrics. (#11868)

🧰 Bug fixes 🧰

  • spanmetricsprocessor: Fix panic caused by race condition when accessing span attributes. (#12644)
  • awsxrayexporter: Stop dropping exception in aws xray events for non error codes (#12643)
  • signalfxexporter: use azure.vm.name instead of host.name to build azure resource id (#12779)
  • receiver/hostmetrics: Do not throw scraping errors if conntrack metrics collection is disabled (#12799)
  • kubeletstatsreceiver: Fetch metadata from initContainers (#12887)
  • metricstransformprocessor: Fix logic in merging exponential histogram. (#12865)
  • resourcedetectionprocessor: Add back host.name attribute when running on GKE (#12354)
  • filelogreceiver: Fix issue where checkpoints could be ignored if start_at`` was set to end`` (#12769)
  • pkg/stanza: Stop readerFactory from returning an error when creating an unsafeReader (#12766) This bug caused the filelog receiver to crash when the collector was restarted and the logs were being read from the end of the file
  • prometheusreceiver: Fix handling of timestamps to prevent reset when a new datapoint is recorded (#12746)
  • exporter/prometheusremotewrite: Fix a panic when a histogram does not have any buckets (#12777)
  • signalfxexporter: fix invalid response error message (#12654)
  • skywalkingreceiver: Add extra link attributes from skywalking ref. (#12651)
  • spanmetricsprocessor: Modifies spanmetrics processor to handle negative durations without crashing. Related to open-telemetry/opentelemetry-js-contrib#1013 (#7250) Sets negative durations to count towards the smallest histogram bucket.
  • splunkhecexporter: use proper config flags to configure content length of gzip buffers for metrics and traces (#12906)

v0.57.1

This version has been skipped.

v0.57.0

This version has been skipped.

v0.56.0

🛑 Breaking changes 🛑

  • datadogexporter: Change default value of host_metadata::hostname_source to config_or_system (#10424)

    • This behavior can be reverted by disabling the exporter.datadog.hostname.preview feature gate.

  • datadogexporter: Make automatically detected hostname match the one reported by other Datadog products. (#10424)

    • This behavior can be reverted by disabling the exporter.datadog.hostname.preview feature gate.

  • dynatraceexporter: Improve serialization of certain Instrument/Temporality pairings (#11828)

    • Non-monotonic cumulative Sum metrics will be exported as Gauges from now on. Non-monotonic cumulative Sums (usually UpDownCounters) will show the current value instead of a change rate.

  • transformprocessor: Field metric.type is now accessed as an enum/int64 instead of a string. (#10349)

  • memcachedreceiver: Remove direction for metrics. The feature gate: receiver.memcachedreceiver.emitMetricsWithoutDirectionAttribute can be set to apply the following (#12404) (#12165)

    • memcached metrics:
      • memcached.network will become:
        • memcached.network.sent
        • memcached.network.received

  • tracegen: Moving component under cmd for consistency (#12474)

  • datadogexporter: Remove deprecated configuration features. (#9099, #9016, #8845, #8783, #8781, #8489, #8396)

  • datadogexporter: Remove deprecated config package. (#8373)

  • observiqexporter: Remove the observiq exporter (#12406)

  • vcenterreceiver: Adds the vcenter.cluster.name resource attribute to vcenter.datastore metrics (#12357)

  • vcenterreceiver: remove metadata declaration and references to uncollected metric vcenter.vm.cpu.utilization (#12358)

  • vcenterreceiver: Remove direction for metrics. The feature gate: receiver.vcenterreceiver.emitMetricsWithoutDirectionAttribute can be set to apply the following (#11849) (#12185)

    • vcenter.host.disk.throughput will become:
      • vcenter.host.disk.throughput.read
      • vcenter.host.disk.throughput.write
    • vcenter.host.disk.latency.avg will become:
      • vcenter.host.disk.latency.avg.read
      • vcenter.host.disk.latency.avg.write
    • vcenter.host.network.throughput will become:
      • vcenter.host.network.throughputt.receive
      • vcenter.host.network.throughput.transmit
    • vcenter.host.network.packet.errors will become:
      • vcenter.host.network.packet.errors.receive
      • vcenter.host.network.packet.errors.transmit
    • vcenter.host.network.packet.count will become:
      • vcenter.host.network.packet.count.receive
      • vcenter.host.network.packet.count.transmit
    • vcenter.vm.disk.latency.avg.read will become:
      • vcenter.vm.disk.latency.avg.read
      • vcenter.vm.disk.latency.avg.write
    • vcenter.vm.network.throughput will become:
      • vcenter.vm.network.throughput.receive
      • vcenter.vm.network.throughput.transmit
    • vcenter.vm.network.packet.count will become:
      • vcenter.vm.network.packet.count.receive
      • vcenter.vm.network.packet.count.transmit

🚩 Deprecations 🚩

  • hostmetricsreceiver: Remove direction for disk metrics. The feature gate: receiver.hostmetricsreceiver.emitMetricsWithoutDirectionAttribute can be set to apply the following (#11849) (#11816)
    • disk scraper metrics:
      • system.disk.io will become:
        • system.disk.io.read
        • system.disk.io.write
      • system.disk.operations will become:
        • system.disk.operations.read
        • system.disk.operations.write
      • system.disk.operation_time will become:
        • system.disk.operation_time.read
        • system.disk.operation_time.write
      • system.disk.merged will become:
        • system.disk.merged.read
        • system.disk.merged.write

🚀 New components 🚀

  • telemetryquerylanguage: Expose the telemetry query language as a package. (#11751)
  • chronyreceiver: -| This component is a pure go implementation for capturing data from chrony (#11789)
  • otlpjsonfilereceiver: Add a new file receiver reading JSON-encoded OTLP data, after serialization specification (#10836)
  • pulsarexporter: Add Apache Pulsar exporter (#9795)
  • solacereceiver: Add Solace receiver to receive trace data from a Solace PubSub+ Event Broker. (#10572)

💡 Enhancements 💡

  • tanzuobservabilityexporter: Improve algorithm to translate OTEL delta exponential histograms into tanzu observability histograms (#12173)

  • internal/scrapertest: Add sortation functions for scrapertest (#10837)

  • elasticsearchreceiver: Add additional metrics (#12176) Add additional metrics for circuit breakers, disk IO r/w, translog, and CPU load.

  • supported platforms: Add linux-ppc64le architecture to cross build tests in CI (#12350)

  • dockerobserver: incorporate observer.EndpointsWatcher in preparation of multiple event subscribers and use existing internal event loop watcher (#10830, #11541)

  • hostmetrics: Adding connection tracking count and max metrics for linux (#11769)

  • hostmetricsreceiver: New config setting scrape_process_delay is used to indicate the minimum amount of time a process must be running before process metrics can be scraped for it. The default value is 0 seconds ("0s"). (#8976)

  • transformprocessor: Add ability to interact with enums via sybmols. (#10349)

  • telemetryquerylanguage: Adds to the grammar the ability to interpret Enums. (#11751)

  • transformprocessor: Update the transform processor to use the Telemetry Query Language package (#11751)

  • transformprocessor: Add delete_key and delete_matching_keys which allow deleting keys from maps. (#11823)

  • receiver/jaeger: Handle spans with own process (#10186)

  • k8sobserver: incorporate observer.EndpointsWatcher in preparation for multiple event subscribers (#10830, #11544)

  • exporter/loki: Handle multi-tenant use-cases (#3121)

  • metricstransformprocessor: Do not produce empty metrics (#12210)

  • schemaprocessor: renaming some internal packages to help improve maintainability (#8495)

  • groupbytrace: Promote groupbytrace to beta (#12169)

  • redisreceiver: Add redis.version resource attribute. (#12139)

  • podmanreceiver: Fetch containers stats one by one and add container image as metric attribute (#9013)

  • all: Components now report their stability level. (#12104)

  • sqlqueryreceiver: Add Oracle DB Support (#12137)

  • sqlserverreceiver: Added tests to validate counter names and a scraper test. (#9374)

  • templatequerylanguage: TQL grammar now supports complex where expressions with and, or and parentheses (#10195)

  • transformprocessor: Add .string accessor to get hex string for trace_id and span_id (#11555)

  • pkg/translator/jaeger: Handle spans with own process (#10186)

🧰 Bug fixes 🧰

  • cmd/chloggen: Compare changelog to common ancestor with main (#12149)
  • logstransformprocessor: Remove support for storage (#12424)
  • elasticsearchreceiver: Remove unused disk_usage_state attribute from documentation (#12429)
  • filterlog: change OR to AND logic for filtering logs - as desribed, and as is done for span filtering (#11439)
  • lokiexporter: Wrap quotes around log message (#11827)
  • tooling: Fix a bug in the makefile causing make rpm-package to fail (#12162)
  • prometheusexporter: Fix cumulative condition for the delta-to-cumulative (#4153)
  • xrayreceiver: Ensure that the UDP Poller ends the obsreport span it creates with every loop. (#12299) This avoids a memory leak that can happen when spans are created to track received segments but never ended.
  • googlecloudexporter: Various bug fixes in parsing OTel logs to GCP LogEntries (#12157)
  • datadogexporter: Fix logs related to the source provider. (#12160)
  • pkg/stanza/fileconsumer: Fix issue where reader could become stuck on newlines (#10125, #10127, #10128)
  • receivercreator: dynamically created receivers log with their name field (#16481)
  • skywalkingreceiver: Fix skywalking traceid and spanid convertion (#11562)
  • spanmetricsprocessor: Fixes the number of explicit bucket counts by removing the manually added "catch-all" bucket. (#11784)
  • spanmetricsprocessor: Fix concurrency bug causing premature key eviction. (#9018)
  • spanmetricsprocessor: Removes a comment that is no longer relevant due to a fix. (#12427)
  • sqlqueryreceiver: Metrics did not contain a timestamp. Timestamps are required for a variety of backends, especially prometheus. In order to work with prometheus we must set the metrics timestamp because metrics with the default timestamp (year 1970) will be dropped by prometheus backends. (#12088)
  • statsdreceiver: Fixing parsing issue for values on statsd receiver (#12120)
  • vcenterreceiver: Fixes calculation of vcenter.vm.disk.utilization. (#12342)

v0.55.0

🛑 Breaking changes 🛑

  • datadogexporter: Remove Sanitize method from Config struct. (#8373)

  • datadogexporter: (Under exporter.datadog.hostname.preview feature gate) Remove docker hostname detector (#11834)

  • k8sclusterreceiver: The receiver.k8sclusterreceiver.reportCpuMetricsAsDouble feature gate has been removed (#10838)

    • If users were disabling this feature gate, they may have to update monitoring for a few Kubernetes cpu metrics. For more details see feature-gate-configurations.

  • prometheusexporter: Automatically rename metrics with units to follow Prometheus naming convention (#8950)

🚩 Deprecations 🚩

  • datadogexporter: Deprecate config package in favor of structs on the datadogexporter package. (#8373)

  • k8sattributesprocessor: Announcing pod_association rules will be deprecated. Use pod_association.sources rules instead. This is in order to support metadata enrichment based on multiple attributes. (#4309)

  • hostmetricsreceiver: Remove direction for network, paging and process metrics. The feature gates receiver.hostmetricsreceiver.emitMetricsWithoutDirectionAttribute and receiver.hostmetricsreceiver.emitMetricsWithDirectionAttribute can be used to control the transition (#11815)

    • system.network.dropped will become:
      • system.network.dropped.receive
      • system.network.dropped.transmit
    • system.network.errors will become:
      • system.network.errors.receive
      • system.network.errors.transmit
    • system.network.io will become:
      • system.network.io.receive
      • system.network.io.transmit
    • system.network.packets will become:
      • system.network.packets.receive
      • system.network.packets.transmit
    • system.paging.operations will become:
      • system.paging.operations.page_in
      • system.paging.operations.page_out
    • process.disk.io will become:
      • process.disk.io.read
      • process.disk.io.write

  • logzioexporter: Announcing custom_endpoint, drain_interval, queue_capacity, queue_max_length configuration options will be deprecated in upcoming releases (#10821)

  • simpleprometheusreceiver: Announcing tls_enable, tls_config will be deprecated and use confighttp.HTTPClientSettings instead. (#11553)

🚀 New components 🚀

  • sqlqueryreceiver: Enable the component (#11848)

💡 Enhancements 💡

  • cmd/chloggen: Update CI process to validate new changelog strategy (#11841)
  • cmd/chloggen: Add new tool for conflict-free CHANGELOG.md management (#11539)
  • coralogixexporter: Add support for metrics (#11065)
  • coreinternal/attraction: Supports pattern for delete and hash attractions (#11886)
  • filterprocessor: Add ability to filter Spans (#6341)
  • flinkmetricsreceiver: Add attribute values to metadata (#11520)
  • k8sattributesprocessor: do not ignore hostNetwork pods for enrichment based on non IP attribute (#12073)
  • logzioexporter: Add support for logs pipeline and support for exporterhelper (#10821)
  • pkg/stanza: Export pkg/stanza/fileconsumer package (#11844)
  • prometheusreceiver: Add target_info labels to resource attributes (#11034)
  • redisreceiver: Add more metrics, redis.maxmemory, redis.role, redis.cmd.calls redis.cmd.usec (#11090)
  • sapmexporter: Add config option to log responses from Splunk APM (#11425)
  • splunkhecexporter: Update limits for max_content_length settings (#11550)
  • sqlqueryreceiver: Add core functionality (#10867)
  • tracegen: Add additional resource attributes (#11145)
  • transformprocessor: Add IsMatch factory function. This function allows regex matching in conditions. (#10903)
  • transformprocessor: replace_patternandreplace_all_patterns` use regex for pattern matching and replacing text in attributes/metrics (#11125)

🧰 Bug fixes 🧰

  • aerospikereceiver: Fix issue where namespaces would not be collected (#11465)
  • aerospikereceiver: Fix typo in metric name. Ensure namespace transactions are collected (#12085) (#12083)
  • coralogixexporter: Fix metrics bearer token (#11831)
  • datadogexporter: (Under exporter.datadog.hostname.preview feature gate) Make the hostname reported on GKE match the Datadog GCP integration hostname. (#11893)
  • datadogexporter: The traces.span_name_remappings setting now correctly refers to the OpenTelemetry key to be renamed without any sort of normalization. (#9693)
  • datadogexporter: Unify traces exporter behavior with Datadog Agent OTLP traces ingest. (#9693)
  • filelogreceiver: Read log lines from lost files first in a poll cycle (#12084)
  • filestorageextension: Copy values returned by Get (#11776)
  • cmd/chloggen: Fix problem where 'new' command would fail when branch name contained '/' (#11887)
  • kafkaexporter: Fixed config.Topic mutation causing Logs|Metrics|Traces to default to 1 topic (#11420)
  • mongodbreceiver: do not ignore TLS Settings in mongodbreceiver (#12092)
  • prometheusexporter: Expose the same metric from different targets correctly (#4986)
  • redactionprocessor: Respect allow_all_keys configuration (#11542)
  • saphanareceiver: Fix component memory query, add better error handling (#11507)
  • sapmreceiver: Fix issue where component instance use in multiple pipelines leads to start failures (#11518)
  • signalfxreceiver: Fix issue where component instance use in multiple pipelines leads to start failures (#11513)
  • splunkhecreceiver: Fix issue where component instance use in multiple pipelines leads to start failures (#11517)

v0.54.0

🛑 Breaking changes 🛑

  • transformprocessor: metric.is_monotonic is now accessed via a bool literal instead of a string. (#10473)
  • vcenterreceiver: Changed the attribute effective on vcenter.cluster.host.count as it will now be reported as a bool rather than a string (#10914)

🚩 Deprecations 🚩

  • datadogexporter: Deprecate instrumentation_library_metadata_as_tags (#11135)
  • datadogexporter: Deprecate Sanitize method of Config struct (#8829)
  • observiqexporter: Deprecate the observiq exporter (#10977)
  • honeycombexporter: Deprecate honeycomb exporter (#10318)

🚀 New components 🚀

  • expvarreceiver: Include expvarreceiver in components (#10847)
  • googlemanagedprometheusexporter Add the Google Managed Service for Prometheus exporter. (#10840)
  • googlemanagedprometheusexporter The Google Managed Service for Prometheus exporter is alpha. (#10925)

💡 Enhancements 💡

  • tailsamplingprocessor: Add trace_state policy (#10852)
  • mongodbatlasreceiver Add support for receiving alerts (#10854)
  • cmd/mdatagen: Allow attribute values of any types (#9245)
  • metricstransformprocessor: Migrate the processor from OC to pdata (#10817)
    • This behavior can be reverted by disabling the processor.metricstransformprocessor.UseOTLPDataModel feature gate.
  • transformprocessor: Add byte slice literal to the grammar. Add new SpanID and TraceID functions that take a byte slice and return a Span/Trace ID. (#10487)
  • transformprocessor: Add Summary transform functions. (#11041)
  • transformprocessor: Add nil literal to the grammar. (#11150)
  • elasticsearchreceiver: Add integration test for elasticsearch receiver (#10165)
  • tailsamplingprocessor: New sampler added that allows to sample based on minimum number of spans
  • datadogexporter: Some config validation and unmarshaling steps are now done on Validate and Unmarshal instead of Sanitize (#8829)
  • datadogexporter: Add exporter.datadog.hostname.preview feature flag and related warnings (#10926)
  • datadogexporter: Add instrumentation_scope_metadata_as_tags instead of instrumentation_library_metadata_as_tags in favor of https://github.com/open-telemetry/opentelemetry-proto/releases/tag/v0.15.0 (#11135)
  • examples: Add an example for scraping Couchbase metrics (#10894)
  • filestorageextension: Add background compaction capability (#9327)
  • googlecloudpubsubreceiver: Added new Endpoint and Insecure connection configuration options. (#10845)
  • dynatraceexporter: Provide better estimated summaries for partial histograms. (#11044)
  • mongodbreceiver: Add integration test for mongodb receiver (#10864)
  • mezmoexporter: add logging for HTTP errors (#10875)
  • googlecloudexporter: Support writing to multiple GCP projects by setting the gcp.project.id resource attribute, and support service account impersonation (#11051)
  • k8sattributeprocessor: Add debug logs to help identify missing attributes (#11060)
  • jmxreceiver: Add latest releases of jmx metrics gatherer & wildfly jar to supported jars hash list (#11134)
  • rabbitmqreceiver: Add integration test for rabbitmq receiver (#10865)
  • transformprocessor: Allow using trace_state with key-value struct (#11029)

🧰 Bug fixes 🧰

  • kubletetstatsreceiver: Bring back k8s.container.name attribute (#10848)
  • transformprocessor: Fix issue where some metric fields were not working correctly in conditions. (#10473)
  • transformprocessor: Fix issue where some trace fields were not working correctly in conditions. (#10471)
  • transformprocessor: Fix issue where some log fields were not working correctly in conditions. (#10903)
  • pkg/stanza: Skip building fingerprint in case of configuration change (#10485)
  • windowseventlogreceiver: Fixed example config in readme (#10971)
  • pkg/stanza: Fix access to atomic variable without using atomic package (#11023)
  • exporter/awsemfexporter:: Fix dead links in README.md. (#11027)
  • googlecloudexporter: Fix (self-obs) point_count metric calculation, concurrent map write panic, and dropped log attributes (#11051)
  • signalfxexporter: Event Type is a required field, if not set, set it to unknown to prevent signalfx ingest from dropping it (#11121)
  • prometheusreceiver: validate that combined metric points (e.g. histograms) have the same timestamp (#9385)
  • splunkhecexporter: Fix flaky test when exporting traces (#11418)
  • mongodbatlasexporter: Fix mongodbatlas.system.memory.usage.max not being reported (#11126)
  • receiver/awsxrayreceiver: Fix null span exception fields causing null pointer exception (#11431)
  • pkg/stanza: use ObservedTimestamp to decide if flush log for recombine operator (#11433)

v0.53.0

🛑 Breaking changes 🛑

  • jmxreceiver: Remove properties & groovyscript parameters from JMX Receiver. Add ResourceAttributes & LogLevel parameter to supply some of the removed functionality with reduced attack surface (#9685)
  • resourcedetectionprocessor: 'gke' and 'gce' resource detectors are replaced with a single 'gcp' detector (#10347)
  • pkg/stanza: Removed reference to deprecated ClusterName (#10426)
  • couchbasereceiver: Fully removed unimplemented Couchbase receiver (#10482)
  • hostmetricsreciever: Fix Load Scraper to normalize 1m, 5m, and 15m averages independently (#8267)

🚀 New components 🚀

  • flinkmetricsreceiver: Add implementation of Flink Metric Receiver (#10121)
  • windowseventlogreceiver Added implementation of Windows Event Log Receiver (#9228)
  • vcenterreceiver: Add metrics receiver for new vcenterreceiver component (#9224)
  • googlecloudpubsubreceiver Activate the Google Cloud Pubsub receiver. (#10580)
  • googlecloudpubsubexporter Activate the Google Cloud Pubsub exporter. (#10580)
  • aerospikereceiver: Add implementation of Aerospike Metric Receiver. (#9961)

💡 Enhancements 💡

  • awsemfexporter: Add min and max support for histograms (#10577)
  • tailsamplingprocessor: Add support for string invert matching to and policy (#9553)
  • mezemoexporter: Add user agent string to outgoing HTTP requests (#10470)
  • prometheusreceiver: Improve performance of metrics builder (#10546)
  • transformprocessor: Add functions for conversion of scalar metric types (gauge_to_sum and sum_to_gauge) (#10255)
  • dynatraceexporter: Use min and max when provided in a data point for histograms (#10815)
  • dynatraceexporter: Truncate unmarshalable responses to avoid long log lines (#10568)
  • scrapertest: Add IgnoreResourceAttributeValue option to metric comparison (#10828)

🧰 Bug fixes 🧰

  • transformprocessor: Fix issue where incorrect error was returned if a bad path was passed to a function (#10141)
  • tanzuobservabilityexporter: Improve how negative values in exponential histograms are handled. (#10135)
  • dynatraceexporter: Ensure min is always less than or equal to mean and max is always greater or equal to mean for histogram estimation. (#10257)
  • resourcedetectionprocessor: GCP resource detector now properly detects zone/region on GKE (#10347)
  • resourcedetectionprocessor: GCP resource detector no longer fails to detect resource when using workload identity (#10486)
  • tailsamplingprocessor: Fix composite sampler with inverse policy
  • awsprometheusremotewriteexporter: Fix signing of empty request bodies. (#10578)
  • sigv4authextension: Fix signing of empty request bodies. (#10578)
  • prometheusexporter: Converting monotonic Delta to Cumulative sums (#9919)
  • statsdreceiver: Update the lastIntervalTime for Counter metrics (#9919)
  • resourcedetectionprocessor: GCP resource detector now correctly detects region on Google App Engine standard (#10814)
  • apachereceiver: Update units to follow semconv (#10587)

v0.52.0

🛑 Breaking changes 🛑

  • jmxreceiver: Hash the jars provided to JMX Receiver and only allow if they match an approved list (#9687)
  • jmxreceiver: Remove properties & groovyscript parameters from JMX Receiver. Add ResourceAttributes & LogLevel parameter to supply some of the removed functionality with reduced attack surface (#9685)

🚀 New components 🚀

  • aerospikereceiver: Add implementation of Aerospike Metrics Receiver (#9961)
  • bigipreceiver: Add implementation of F5 Big-IP Metric Receiver (#9680)
  • expvarreceiver: Initial work for a receiver designed to scrape memstats from Golang applications. (#9747)
  • mezmoexporter: Add implementation of Mezmo Log exporter (#9743)
  • nsxtreceiver: Added implementation of NSX-T Metric Receiver (#9568)
  • expvarreceiver: Add implementation of new receiver. (#10183)
  • telemetrygen: Started implementing an upgraded version of tracegen generating traces and metrics (#9597)

💡 Enhancements 💡

  • transformprocessor: Add transformation of metrics (#10100)
  • transformprocessor: Include transform processor in components (#10134)
  • kubeletstatsreceiver: Update receiver to use new Metrics Builder. All emitted metrics remain the same. (#9744)
  • transformprocessor: Add new replace_match and replace_all_matches functions (#10132)
  • resourcedetectionprocessor: Add "cname" and "lookup" hostname sources
  • jmxreceiver: Communicate with JMX metrics gatherer subprocess via properties file (#9685)
  • pkg/stanza: make multiline tests more like integration tests #10353

🧰 Bug fixes 🧰

  • datadogexporter: add error checks for datadog exporter (#9964)
  • datadogexporter: Fix host aliases not being properly sent to the Datadog backend (#9748)
  • groupbyattrsprocessor: copied aggregationtemporality when grouping metrics. (#9088)
  • jaeger: Update OTLP-Jaeger translation of span events according to the OTel Spec: use event log field instead of message to represent OTel Span Event Name (#10273)
  • mongodbreceiver: Fix issue where receiver startup could hang (#10111)
  • transformprocessor: Fix issue where metric.aggregation_temporality and metric.is_monotic were not actually gettable or settable (#10197)
  • signalfxexporter: Emit prometheus compatible histogram/summary to signalfx #10299
    • This behavior can be reverted using the exporter.signalfxexporter.PrometheusCompatible featuregate.
  • podmanreceiver: Container Stats Error structure (#9397)
  • pkg/stanza: pipeline.Operators() will return a consistently ordered list of operators whenever possible (#9761)
  • tanzuobservabilityexporter: add error checks for tanzuobservability exporter (#10188)

v0.51.0

🛑 Breaking changes 🛑

  • datadogexporter: Replace HistogramMode defined as string with enum. (#9589)
  • pkg/translator/signalfx: Change signalfx translator to expose To/From translator structs. (#9740)
  • transformprocessor: Add parameter validation to truncate_all and limit functions. The limit parameter can no longer be negative. (#9783)
  • newrelicexporter deleted. Use New Relic native OTLP ingest instead. (#9894)
  • k8sclusterreceiver: Removing ClusterName as per https://github.com/kubernetes/apimachinery/commit/430b920312ca0fa10eca95967764ff08f34083a3. (#9885)

🚩 Deprecations 🚩

  • exporter/azuremonitor: Deprecate use of LogRecord.Name as the log envelope category name. There is no replacement. (#9258)
  • processor/k8sattributes: Deprecate use of k8s.cluster.name metadata parameter (obsolete) (#9968)

🚀 New components 🚀

  • schemaprocessor: Starting the initial work to allow from translating from semantic convention to another (#8371)
  • saphanareceiver: Added implementation of SAP HANA Metric Receiver (#8827)
  • logstransformprocessor: Add implementation of Logs Transform Processor (#9335)

💡 Enhancements 💡

  • cmd/mdatagen: Replace enum attributes values with typed constants (#9683)
  • elasticsearchreceiver: Update metrics scope name from otelcol/elasticsearch to otelcol/elasticsearchreceiver (#9757)
  • k8sclusterreceiver: Validate that k8s API supports a resource before setting up a watcher for it (#9523)
  • internal/stanza: Add support for remove operator (#9524)
  • k8sattributesprocessor: Support regex capture groups in tag_name (#9525)
  • mongoreceiver: Update metrics scope name from otelcol/mongodb to otelcol/mongodbreceiver (#9759)
  • transformprocessor: Add new truncation function to allow truncating string values in maps such as attributes or resource.attributes (#9546)
  • datadogexporter: Add api.fail_on_invalid_key to fail fast if api key is invalid (#9426)
  • transformprocessor: Add support for functions to validate parameters (#9563)
  • googlecloudexporter: Add GCP cloud logging exporter (#9679)
  • transformprocessor: Add new limit function to allow limiting the number of items in a map, such as the number of attributes in attributes or resource.attributes (#9552)
  • processor/attributes: Support attributes set by server authenticator (#9420)
  • datadogexporter: Experimental support for Exponential Histograms with delta aggregation temporality (#8350)
  • prometheusreceiver: Support OpenMetrics Info and Stateset metrics (#9378)

🧰 Bug fixes 🧰

  • k8sclusterreceiver: Fix the receiver to work with 1.19 and 1.20 k8s API versions (#9523)
  • azuremonitorexporter: Fix log exporter bug related to incorrectly mapping SpanId (#9579)
  • mysqlreceiver: Fix attribute values mismatch with its definition (#9688)
  • opencensusreceiver: Do not report fatal error if err is server closed (#9559).
  • sqlserverreceiver: Fix the receiver to have integer types on metrics where applicable (#9601)
  • prometheusreceiver: Fix the memory issue introduced in the 0.49.0 release (#9718)
  • couchdbreceiver: Fix issue where the receiver would not respect custom metric settings (#9598)
  • nginxreceiver: Include nginxreceiver in components (#9572)
  • pkg/translator/prometheusremotewrite: Fix data race when used with other exporters (#9736)
  • examples/demo: fix baggage not work in trace demo app. (#9418)
  • prometheusreceiver: Handle the condition where up metric value is NaN (#9253)
  • tanzuobservabilityexporter: Make metrics stanza in config be optional (#9098)
  • filelogreceiver: Update Kubernetes examples to fix native OTel logs collection issue where 0 length logs cause errors (#9754)
  • logstransformprocessor: Resolve node ordering to fix intermittent failures (#9761)
  • awsinsightreceiver: Migrate from ConfigMapsResourceLock to ConfigMapsLeasesResourceLock as per https://github.com/kubernetes/client-go/commit/276ea3ed979947d7cdd4b3d708862245ddcd8883 (#9885)
  • filelog, journald, syslog, tcplog, udplog: Add support for []string type for converting log record entries (#9887)

v0.50.0

🛑 Breaking changes 🛑

  • stackdriverexporter: Remove the stackdriver exporter in favor of the identical googlecloud exporter (#9274)
  • filelog, journald, syslog, tcplog, udplog: Remove preserve_to field from sub-parsers (#9331)
  • kafkametricsreceiver: instrumentation name updated from otelcol/kafkametrics to otelcol/kafkametricsreceiver (#9406)
  • kubeletstatsreceiver: instrumentation name updated from kubeletstats to otelcol/kubeletstatsreceiver (#9400)
  • datadogexporter: Remove GetHostTags method from TagsConfig struct (#9423)
  • googlecloudexporter: Graduate the exporter.googlecloud.OTLPDirect feature-gate to Beta. This includes changes to the configuration structure, and many changes to default behavior. (#9471)

🚩 Deprecations 🚩

  • cumulativetodeltaprocessor: Deprecated metrics configuration option in favor of include and exclude (#8952)
  • datadogexporter: Deprecate metrics::report_quantiles in favor of metrics::summaries::mode (#8846)
  • datadogexporter: Deprecate traces.sample_rate setting. It was never used anywhere. (#9771)

🚀 New components 🚀

  • iisreceiver: Add implementation of IIS Metric Receiver (#8832)
  • sqlserverreceiver: Add implementation of SQL Server Metric Receiver (#8398)
  • activedirectorydsreceiver: Add implementation of Active Directory Domain Services metric receiver (#9359)
  • sqlqueryreceiver: Add readme, factory, and config to initial implementation of SQL receiver (#9408)

💡 Enhancements 💡

  • pkg/translator/prometheusremotewrite: Allow to disable sanitize metric labels (#8270)
  • basicauthextension: Implement configauth.ClientAuthenticator so that the extension can also be used as HTTP client basic authenticator.(#8847)
  • azuremonitorexporter, lokiexporter, observiqexporter: Update timestamp processing logic (#9130)
  • cumulativetodeltaprocessor: add new include/exclude configuration options with regex support (#8952)
  • datadogexporter: Update deprecation messages to reflect new deprecation plan (#9422)
  • cmd/mdatagen: Update generated functions to have simple parse function to handle string parsing consistently and limit code duplication across receivers (#7574)
  • attributesprocessor: Support filter by severity (#9132)
  • transformprocessor: Add transformation of logs (#9368)
  • datadogexporter: Add metrics::summaries::mode to specify export mode for summaries (#8846)
  • prometheusreceiver: Add resource attributes for kubernetes resource discovery labels (#9416)

🧰 Bug fixes 🧰

  • fluentforwardreceiver: Release port on shutdown (#9111)
  • prometheusexporter: Prometheus fails to generate logs when prometheus exporter produced a check exception occurs. (#8949)
  • resourcedetectionprocessor: Wire docker detector (#9372)
  • kafkametricsreceiver: The kafkametricsreceiver was changed to connect to kafka during scrape, rather than startup. If kafka is unavailable the receiver will attempt to connect during subsequent scrapes until succcessful (#8817).
  • datadogexporter: Update Kubernetes example manifest to new executable name. (#9425).
  • riakreceiver: Fix issue where user configured metric settings were ignored. (#9561)
  • sqlserverreceiver: Update sqlserver.transaction_log.growth.count and sqlserver.transaction_log.shrink.count to be monotonic sums. (#9522)

v0.49.0

⚠️ Warning ⚠️

This release contains an issue in Prometheus receiver causing 30% memory consumption increase when there is a lot of target churn. The issue is currently being investigated and will be fixed in one of the new releases. More details: open-telemetry#9278.

🛑 Breaking changes 🛑

  • filelogreceiver, journaldreceiver, syslogreceiver, tcplogreceiver, udplogreceiver:
    • Updated data model to align with stable logs data model, which includes various breaking changes. (#9139, #8835)
      • A detailed Upgrade Guide is available in the log-collection v0.29.0 release notes.
  • datadogexporter: Remove OnlyMetadata method from Config struct (#8980)
  • datadogexporter: Remove GetCensoredKey method from APIConfig struct (#8980)
  • mongodbatlasreceiver: Updated to uses newer metric builder which changed some metric and resource attributes (#9093)
  • dynatraceexporter: Make serialization package /internal (#9097)
  • attributesprocessor: Remove log names from filters (#9131)
  • k8sclusterreceiver: The receiver.k8sclusterreceiver.reportCpuMetricsAsDouble feature gate is now enabled by default (#9367)

🚩 Deprecations 🚩

  • datadogexporter: Deprecate service setting in favor of service.name semantic convention (#8784)
  • datadogexporter: Deprecate version setting in favor of service.version semantic convention (#8784)
  • datadogexporter: Deprecate env setting in favor of deployment.environment semantic convention (#9017)
  • datadogexporter: Deprecate GetHostTags method from TagsConfig struct (#8975)
  • datadogexporter: Deprecate tags setting in favor of host_metadata::tags (#9100)
  • datadogexporter: Deprecate send_metadata setting in favor of host_metadata::enabled (#9100)
  • datadogexporter: Deprecate use_resource_metadata setting in favor of host_metadata::hostname_source (#9100)
  • prometheusexecreceiver: Deprecate prom_exec receiver (#9058)
  • fluentbitextension: Deprecate Fluentbit extension (#9062)

🚀 New components 🚀

  • riakreceiver: Riak Metric Receiver (#8548)

💡 Enhancements 💡

  • splunkhecexporter: Add support for batching traces (#8995)
  • hostmetricsreceiver: Migrate Processes scraper to the Metrics builder (#8855)
  • tanzuobservabilityexporter: Use resourcetotelemetry helper (#8338)
  • Add make crosslink target to ensure replace statements are included in go.mod for all transitive dependencies within repository (#8822)
  • filestorageextension: Change bbolt DB settings for better performance (#9004)
  • jaegerremotesamplingextension: Add local and remote sampling stores (#8818)
  • attributesprocessor: Add support to filter on log body (#8996)
  • prometheusremotewriteexporter: Translate resource attributes to the target info metric (#8493)
  • prometheusexporter: Add job and instance labels to metrics so they can be scraped with honor_labels: true (#9115)
  • podmanreceiver: Add API timeout configuration option (#9014)
  • cmd/mdatagen: Add sem_conv_version field to metadata.yaml that is used to set metrics SchemaURL (#9010)
  • splunkheceporter: Add an option to disable log or profiling data (#9065)
  • windowsperfcountersreceiver: Move code into separate package for use in other windowsperfcounter receivers (#9108)
  • datadogexporter: Add host_metadata configuration section to configure host metadata export (#9100)
  • cmd/mdatagen: Update documentation generated for attributes to list enumerated values and show the "value" that will be visible on metrics when it is different from the attribute key in metadata.yaml (#8983)
  • routingprocessor: add option to drop resource attribute used for routing (#8990)

🧰 Bug fixes 🧰

  • filestorageextension: use correct bbolt options for compaction (#9134)
  • hostmetricsreceiver: Use cpu times for time delta in cpu.utilization calculation (#8857)
  • dynatraceexporter: Remove overly verbose stacktrace from certain logs (#8989)
  • googlecloudexporter: fix the exporter.googlecloud.OTLPDirect fature-gate, which was not applied when the flag was provided (#9116)
  • signalfxexporter: Fix bug to enable timeouts for correlating traces and metrics (#9101)
  • windowsperfcountersreceiver: fix exported values being integers instead of doubles (#9138)
  • prometheusreceiver: Fix issues with relabelling the job and instance labels. (#8780)
  • dynatraceexporter: Continue processing data points after a serialization error. (#9330)

v0.48.0

💡 Enhancements 💡

  • k8seventsreceiver: Add Api_version and resource_version (#8539)
  • datadogexporter: Add metrics::sums::cumulative_monotonic_mode to specify export mode for cumulative monotonic sums (#8490)
  • dynatraceexporter: add multi-instance deployment note to README.md (#8848)
  • resourcedetectionprocessor: Add attribute allowlist (#8547)
  • datadogexporter: Metrics payload data and Sketches payload data will be logged if collector is started in debug mode (#8929)
  • cmd/mdatagen: Add resource attributes definition to metadata.yaml and move pdata.Metrics creation to the generated code (#8555)

🛑 Breaking changes 🛑

  • windowsperfcountersreceiver: Added metrics configuration (#8376)
  • lokiexporter: Remove deprecated LogRecord.name field (#8951)
  • splunkhecexporter: Remove deprecated LogRecord.name field (#8951)

🚩 Deprecations 🚩

  • datadogexporter: Deprecate OnlyMetadata method from Config struct (#8359)
  • datadogexporter: Deprecate GetCensoredKey method from APIConfig struct (#8830)
  • datadogexporter: Deprecate metrics::send_monotonic_counter in favor of metrics::sums::cumulative_monotonic_mode (#8490)

🚀 New components 🚀

  • sigv4authextension: Enable component (#8518)

v0.47.0

💡 Enhancements 💡

  • googlecloudexporter: Add Validate method in config (#8559)
  • attributesprocessor: Add convert action (#7930)
  • attributesprocessor: Add metric support (#8111)
  • prometheusremotewriteexporter: Write-Ahead Log support enabled (#7304)
  • hostreceiver/filesystemscraper: Add filesystem utilization (#8027)
  • hostreceiver/pagingscraper: Add paging.utilization (#6221)
  • googlecloudexporter: [Alpha] Translate metrics directly from OTLP to gcm using the exporter.googlecloud.OTLPDirect feature-gate (#7177)
  • simpleprometheusreceiver: Add support for static labels (#7908)
  • spanmetricsprocessor: Dropping the condition to replace _ with key_ as __ label is reserved and _ is not (#8057)
  • podmanreceiver: Add container.runtime attribute to container metrics (#8262)
  • dockerstatsreceiver: Add container.runtime attribute to container metrics (#8261)
  • tanzuobservabilityexporter: instrumentation Library and Dropped Counts to Span Tags (#8120)
  • clickhouseexporter: Implement consume log logic. (#9705)
  • influxdbexporter: Add support for cumulative, non-monotonic metrics. (#8348)
  • oauth2clientauthextension: Add support for EndpointParams (#7307)
  • Add NewMetricData function to MetricsBuilder to consistently set instrumentation library name (#8255)
  • googlecloudpubsubreceiver Added implementation of Google Cloud Pubsub receiver. (#8391)
  • googlecloudpubsubexporter Added implementation of Google Cloud Pubsub exporter. (#8391)
  • coralogixexporter Allow exporter timeout to be configured (#7957)
  • prometheusremotewriteexporter support adding trace id and span id attached to exemplars (#8380)
  • influxdbexporter: accept histogram metric missing infinity bucket. (#8462)
  • skywalkingreceiver: Added implementation of Skywalking receiver. (#8549)
  • prometheusreceiver: Fix staleness bug for histograms and summaries (#8561)

🛑 Breaking changes 🛑

  • mongodbatlasreceiver: rename mislabeled attribute memory_state to correct disk_status on partition disk metrics (#7747)
  • mongodbatlasreceiver: Correctly set initial lookback for querying mongodb atlas api (#8246)
  • nginxreceiver: instrumentation name updated from otelcol/nginx to otelcol/nginxreceiver (#8255)
  • postgresqlreceiver: instrumentation name updated from otelcol/postgresql to otelcol/postgresqlreceiver (#8255)
  • redisreceiver: instrumentation name updated from otelcol/redis to otelcol/redisreceiver (#8255)
  • apachereceiver: instrumentation name updated from otelcol/apache to otelcol/apachereceiver ()
  • couchdbreceiver: instrumentation name updated from otelcol/couchdb to otelcol/couchdbreceiver (#8366)
  • prometheusreceiver Change resource attributes on metrics: instance -> service.instance.id, host.name -> net.host.name, port -> net.host.port, scheme -> http.scheme, job removed (#8266)
  • prometheusremotewriteexporter Use service.* resource attributes instead of job and instance resource attributes when adding job and instance labels to metrics (#8266)
  • mysqlreceiver: instrumentation name updated from otel/mysql to otelcol/mysqlreceiver (#8387)
  • zookeeperreceiver: instrumentation name updated from otelcol/zookeeper to otelcol/zookeeperreceiver (#8389)
  • coralogixexporter: Create dynamic subsystem name (#7957)
    • Deprecate configuration changed. Dynamic subsystem name from traces service name property.
  • rabbitmqreceiver: instrumentation name updated from otelcol/rabbitmq to otelcol/rabbitmqreceiver (#8400)

🧰 Bug fixes 🧰

  • zipkinexporter: Set "error" tag value when status is set to error (#8187)
  • prometheusremotewriteexporter: Correctly handle metric labels which collide after sanitization (#8378)
  • prometheusremotewriteexporter: Drop labels when exemplar attributes exceed the max number of characters (#8379)
  • k8sclusterreceiver: Add support to enable k8s node and container cpu metrics to be reported as double values (#8245)
    • Use "--feature-gates=receiver.k8sclusterreceiver.reportCpuMetricsAsDouble" to enable reporting node and container cpu metrics as a double values.
  • tanzuobservabilityexporter: Fix a typo in Instrumentation Library name and version tags (#8384)
  • logreceivers: Fix an issue where receiver would sometimes fail to build using Go 1.18 (#8521)
  • awsxrayreceiver: Add defaults for optional stack frame parameters (#8790)

🚩 Deprecations 🚩

  • datadogexporter: Deprecate automatic environment variable detection (#8397)

🚀 New components 🚀

  • sigv4authextension: New Component: Sigv4 Authenticator Extension (#8263)

v0.46.0

💡 Enhancements 💡

  • internal/stanza: Export metrics from Stanza receivers (#8025)
  • hostreceiver/pagingscraper: Migrate the scraper to the mdatagen metrics builder (#7139)
  • Do not drop zero trace/span id spans in the jaeger conversion (#7946)
  • Upgrade to use semantic conventions 1.6.1 (#7926)
  • dynatraceexporter: Validate QueueSettings and perform config validation in Validate() instead (#8020)
  • sapmexporter: Add validation for sending_queue setting (#8023)
  • signalfxexporter: Add validation for sending_queue setting (#8026)
  • internal/stanza: Add support for arbitrary attribute types (#8081)
  • resourcedetectionprocessor: Add confighttp.HTTPClientSettings To Resource Detection Config Fixes (#7397)
  • hostmetricsreceiver: Add cpu.utilization metrics to cpu scrapper (#7130)
  • honeycombexporter: Add validation for sending_queue setting (#8113)
  • routingprocessor: Expand error handling on failure to build exporters (#8125)
  • skywalkingreceiver: Add new skywalking receiver component folder and structure (#8107)
  • groupbyattrsprocesor: Allow empty keys, which allows to use the processor for compaction (#7793)
  • datadogexporter: Add rbac to example k8s manifest file (#8186)
  • splunkhecexporter: Add validation for sending_queue setting (#8256)

🛑 Breaking changes 🛑

  • Remove deprecated functions from jaeger translator (#8032)
  • internal/stanza: Remove write_to setting from input operators (#8081)
  • mongodbatlasreceiver: rename mongodb.atlas.* attributes to mongodb_atlas.* adhering to naming guidelines. Adding 3 new attributes (#7960)

🧰 Bug fixes 🧰

  • prometheusreceiver: Fix segfault that can occur after receiving stale metrics (#8056)
  • filelogreceiver: Fix issue where logs could occasionally be duplicated (#8123)
  • prometheusremotewriteexporter: Fix empty non-string resource attributes (#8116)

🚀 New components 🚀

v0.45.1

💡 Enhancements 💡

  • sumologicexporter: Move validation to Config (#7936)
  • elasticsearchexporter: Fix crash with batch processor (#7953).
  • splunkhecexporter: Batch metrics payloads (#7760)
  • tanzuobservabilityexporter: Add internal SDK metric tag (#7826)
  • hostreceiver/processscraper: Migrate the scraper to the mdatagen metrics builder (#7287)

🧰 Bug fixes 🧰

  • awsprometheusremotewriteexporter: fix dependencies issue (#7963)

🚀 New components 🚀

  • awsfirehose receiver: Add AWS Kinesis Data Firehose Receiver (#7918)

v0.45.0

💡 Enhancements 💡

  • hostreceiver/filesystemscraper: Migrate the scraper to the mdatagen metrics builder (#7772)
  • hostreceiver/memoryscraper: Migrate the scraper to the mdatagen metrics builder (#7312)
  • lokiexporter: Use record attributes as log labels (#7569)
  • routingprocessor: Do not err on failure to build exporters (#7423)
  • apachereceiver: Update to mdatagen v2 (#7573)
  • datadogexporter: Don't send host metadata if hostname is empty (#7426)
  • datadogexporter: Add insecure_skip_verify flag to configuration (#7422)
  • coralogixexporter: Update readme (#7785)
  • awscloudwatchlogsexporter: Remove name from aws cloudwatch logs exporter (#7554)
  • tanzuobservabilityexporter: Update OTel Collector's Exporter to match WF Proxy Handling of source (#7929)
  • hostreceiver/memoryscraper: Add memory.utilization (#6221)
  • awskinesisexporter: Add Queue Config Validation AWS Kinesis Exporter (#7835)
  • elasticsearchexporter: Remove usage of deprecated LogRecord.Name field (#7829).
  • loadbalancingexporter: Allow non-exist hostname on startup (#7935)
  • datadogexporter: Use exact sum, count and average on Datadog distributions (#7830)
  • storage/filestorage: add optional compaction to filestorage (#7768)
  • tanzuobservabilityexporter: Add attributes from the Resource to the resulting WF metric tags & set source value in WF metric (#8101)

🛑 Breaking changes 🛑

  • Use go mod compat, drops support for reproducibility with go 1.16 (#7915)
  • apachereceiver: Update instrumentation library name from otel/apache to otelcol/apache (#7754)
  • pkg/translator/prometheusremotewrite: Cleanup prw translator public functions (#7776)
  • prometheusreceiver: The OpenCensus-based metric conversion pipeline has been removed.
    • The receiver.prometheus.OTLPDirect feature gate has been removed as the direct pipeline is the only remaining pipeline.
  • translator/jaeger: Cleanup jaeger translator function names (#7775)
    • Deprecate old funcs with Internal word.
  • mysqlreceiver: Update data model and names for several metrics (#7924)
    • Change all metrics to Int values
    • Remove mysql.buffer_pool_pages. Replace with:
      • mysql.buffer_pool.pages
      • mysql.buffer_pool.data_pages
      • mysql.buffer_pool.page_flushes
    • Remove mysql.buffer_pool_size. Replace with:
      • mysql.buffer_pool.limit
      • mysql.buffer_pool.usage
    • Rename mysql.buffer_pool_operations to mysql.buffer_pool.operations

🚩 Deprecations 🚩

  • Deprecated log_names setting from filter processor. (#7552)

🧰 Bug fixes 🧰

  • tailsamplingprocessor: "And" policy only works as a sub policy under a composite policy (#7590)
  • prometheusreceiver: Correctly map description and units when converting Prometheus metadata directly to pdata. (#7748)
  • sumologicexporter: fix exporter panics on malformed histogram (#7548)
  • awsecscontainermetrics: CPU Reserved is now 1024/vCPU for ECS Container Insights (#6734)

🚀 New components 🚀

  • clickhouse exporter: Add ClickHouse Exporter (#6907)
  • pkg/translator/signalfx: Extract signalfx to metrics conversion in a separate package (#7778)
    • Extract FromMetrics to SignalFx translator package (#7823)

v0.44.0

💡 Enhancements 💡

  • kafkaexporter: Add compression and flush max messages options.
  • dynatraceexporter: Write error logs using plugin logger (#7360)
  • dynatraceexporter: Fix docs for TLS settings (#7568)
  • tanzuobservabilityexporter: Turn on metrics exporter (#7281)
  • attributesprocessor resourceprocessor: Add from_context value source
  • resourcedetectionprocessor: check cluster config to verify resource is on aws for eks resources (#7186)
  • awscloudwatchlogsexporter: enable awscloudwatchlogsexporter which accepts and exports log data (#7297)
  • translator/prometheusremotewrite: add a new module to help translate data from OTLP to Prometheus Remote Write (#7240)
  • azuremonitorexporter: In addition to traces, export logs to Azure Application Insights (#7403)
  • jmxreceiver: Added additional_jars configuration option to launch JMX Metric Gatherer JAR with extended CLASSPATH (#7378)
  • awscontainerinsightreceiver: add full pod name when configured to AWS Container Insights Receiver (#7415)
  • hostreceiver/loadscraper: Migrate the scraper to the mdatagen metrics builder (#7288)
  • awsecscontainermetricsreceiver: Rename attributes to follow semantic conventions (#7425)
  • datadogexporter: Always map conventional attributes to tags (#7185)
  • mysqlreceiver: Add golden files for integration test (#7303)
  • nginxreceiver: Standardize integration test (#7515)
  • mysqlreceiver: Update to use mdatagen v2 (#7507)
  • postgresqlreceiver: Add integration tests (#7501)
  • apachereceiver: Add integration test (#7517)
  • mysqlreceiver: Use scrapererror to report errors (#7513)
  • postgresreceiver: Update to mdatagen v2 (#7503)
  • nginxreceiver: Update to mdatagen v2 (#7549)
  • datadogexporter: Fix traces exporter's initialization log (#7564)
  • tailsamplingprocessor: Add And sampling policy (#6910)
  • coralogixexporter: Add Coralogix Exporter (#7383)
  • prometheusexecreceiver: Add default value for scrape_timeout option (#7587)

🛑 Breaking changes 🛑

  • resourcedetectionprocessor: Update os.type attribute values according to semantic conventions (#7544)
  • awsprometheusremotewriteexporter: Deprecation notice; may be removed after v0.49.0
    • Switch to using the prometheusremotewriteexporter + sigv4authextension instead

🧰 Bug fixes 🧰

  • resourcedetectionprocessor: fix meta allow list excluding keys with nil values (#7424)
  • postgresqlreceiver: Fix issue where empty metrics could be returned after failed connection (#7502)
  • resourcetotelemetry: Ensure resource attributes are added to summary and exponential histogram data points. (#7523)

🚩 Deprecations 🚩

  • Deprecated otel_to_hec_fields.name setting from splunkhec exporter. (#7560)

v0.43.0

💡 Enhancements 💡

  • coralogixexporter: First implementation of Coralogix Exporter (#6816)
  • cloudfoundryreceiver: Enable Cloud Foundry client (#7060)
  • elasticsearchexporter: add elasticsearchexporter to the components exporter list (#6002)
  • elasticsearchreceiver: Add metric metadata (#6892)
  • elasticsearchreceiver: Use same metrics as JMX receiver for JVM metrics (#7160)
  • elasticsearchreceiver: Implement scraping logic (#7174)
  • datadogexporter: Add http.status_code tag to trace stats (#6889)
  • datadogexporter: Add configuration option to use OTel span name into the Datatog resource name (#6611)
  • mongodbreceiver: Add initial client code to the component (#7125)
  • tanzuobservabilityexporter: Support delta histograms (#6897)
  • awscloudwatchlogsexporter: Use cwlogs package to export logs (#7152)
  • mysqlreceiver: Add the receiver to available components (#7078)
  • tanzuobservabilityexporter: Documentation for the memory_limiter configuration (#7164)
  • dynatraceexporter: Do not shut down exporter when metrics ingest module is temporarily unavailable (#7161)
  • mongodbreceiver: Add metric metadata (#7163)
  • mongodbreceiver: Add metric scraping (#7175)
  • postgresqlreceiver: add the receiver to available components (#7079)
  • rabbitmqreceiver: Add scraper logic (#7299)
  • tanzuobservability exporter: Support summary metrics (#7121)
  • mongodbatlasreceiver: Add retry and backoff to HTTP client (#6943)
  • Use Jaeger gRPC instead of Thrift in the docker-compose example (#7243)
  • tanzuobservabilityexporter: Support exponential histograms (#7127)
  • receiver_creator: Log added and removed endpoint env structs (#7248)
  • prometheusreceiver: Use the OTLP data conversion path by default. (#7282)
    • Use --feature-gates=-receiver.prometheus.OTLPDirect to re-enable the OpenCensus conversion path.
  • extension/observers: Correctly set image and tag on container endpoints (#7279)
  • tanzuobservabilityexporter: Document how to enable memory_limiter (#7286)
  • hostreceiver/networkscraper: Migrate the scraper to the mdatagen metrics builder (#7048)
  • hostmetricsreceiver: Add MuteProcessNameError config flag to mute specific error reading process executable (#7176)
  • scrapertest: Improve comparison logic (#7305)
  • hostmetricsreceiver: add cpu_average option for load scraper to report the average cpu load (#6999)
  • scrapertest: Add comparison option to ignore specific attributes (#6519)
  • tracegen: Add option to pass in custom headers to export calls via command line (#7308)
  • tracegen: Provide official container images (#7179)
  • scrapertest: Add comparison function for pdata.Metrics (#7400)
  • prometheusremotewriteexporter : Dropping the condition to replace _ with key_ as __ label is reserved and _ is not (#7112)

🛑 Breaking changes 🛑

  • tanzuobservabilityexporter: Remove status.code
  • tanzuobservabilityexporter: Use semantic conventions for status.message (#7126)
  • k8sattributesprocessor: Move kube and observability packages to internal folder (#7159)
  • k8sattributesprocessor: Unexport processor Options (#7311)
  • zookeeperreceiver: Refactored metrics to have correct units, types, and combined some metrics via attributes. (#7280)
  • prometheusremotewriteexporter: PRWExporter struct and NewPRWExporter() function are now unexported. (#TBD)
  • newrelicexporter marked as deprecated (#7284)

🚀 New components 🚀

  • rabbitmqreceiver: Establish codebase for RabbitMQ metrics receiver (#7239)
  • Add basicauth extension (#7167)
  • k8seventsreceiver: Implement core logic (#6885)

🧰 Bug fixes 🧰

  • k8sattributeprocessor: Parse IP out of net.Addr to correctly tag k8s.pod.ip (#7077)
  • k8sattributeprocessor: Process IP correctly for net.Addr instances that are not typed (#7133)
  • mdatagen: Fix validation of enabled field in metadata.yaml (#7166)
  • elasticsearch: Fix timestamp for each metric being startup time (#7255)
  • prometheusremotewriteexporter: Fix index out of range panic caused by expiring metrics (#7149)
  • resourcedetection: Log the error when checking for ec2metadata availability (#7296)

v0.42.0

💡 Enhancements 💡

  • couchbasereceiver: Add couchbase client (#7122)
  • couchdbreceiver: Add couchdb scraper (#7131)
  • couchdbreceiver: Add couchdb client (#6880)
  • elasticsearchreceiver: Implement scraper client (#7019)
  • couchdbreceiver: Add metadata metrics (#6878)
  • prometheusremotewriteexporter: Handling Staleness flag from OTLP (#6679)
  • prometheusexporter: Handling Staleness flag from OTLP (#6805)
  • prometheusreceiver: Set OTLP no-data-present flag for stale scraped metrics. (#7043)
  • mysqlreceiver: Add Integration test (#6916)
  • datadogexporter: Add compatibility with ECS Fargate semantic conventions (#6670)
  • k8s_observer: discover k8s.node endpoints (#6820)
  • redisreceiver: Add missing description fields to keyspace metrics (#6940)
  • redisreceiver: Set start timestamp uniformly for gauge and sum metrics (#6941)
  • kafkaexporter: Allow controlling Kafka acknowledgment behaviour (#6301)
  • lokiexporter: Log the first part of the http body on failed pushes to loki (#6946)
  • resourcedetectionprocessor: add the consul detector (#6382)
  • awsemfexporter: refactor cw_client logic into separate cwlogs package (#7072)
  • prometheusexporter: Dropping the condition to replace _ with key_ as __ label is reserved and _ is not (#7506)

🛑 Breaking changes 🛑

  • memcachedreceiver: Update metric names (#6594)
  • memcachedreceiver: Fix some metric units and value types (#6895)
  • sapm receiver: Use Jaeger status values instead of OpenCensus (#6682)
  • jaeger receiver/exporter: Parse/set Jaeger status with OTel spec values (#6682)
  • awsecscontainermetricsreceiver: remove tag from container.image.name (#6436)
  • k8sclusterreceiver: remove tag from container.image.name (#6436)

🚀 New components 🚀

  • ecs_task_observer: Discover running containers in AWS ECS tasks (#6894)
  • mongodbreceiver: Establish codebase for MongoDB metrics receiver (#6972)
  • couchbasereceiver: Establish codebase for Couchbase metrics receiver (#7046)
  • dbstorage: New experimental dbstorage extension (#7061)
  • redactionprocessor: Remove sensitive data from traces (#6495)

🧰 Bug fixes 🧰

  • ecstaskobserver: Fix "Incorrect conversion between integer types" security issue (#6939)
  • Fix typo in "direction" metrics attribute description (#6949)
  • zookeeperreceiver: Fix issue where receiver could panic during shutdown (#7020)
  • prometheusreceiver: Fix metadata fetching when metrics differ by trimmable suffixes (#6932)
  • Sanitize URLs being logged (#7021)
  • prometheusreceiver: Fix start time tracking for long scrape intervals (#7053)
  • signalfxexporter: Don't use syscall to avoid compilation errors on some platforms (#7062)
  • tailsamplingprocessor: Add support for new policies as composite sub-policies (#6975)

💡 Enhancements 💡

  • lokiexporter: add complete log record to body (#6619)
  • k8sclusterreceiver add container.image.tag attribute (#6436)
  • spanmetricproccessor: use an LRU cache for the cached Dimensions key-value pairs (#2179)
  • skywalkingexporter: add skywalking metrics exporter (#6528)
  • deltatorateprocessor: add int counter support (#6982)
  • filestorageextension: document default values (#7022)
  • redisreceiver: Migrate the scraper to the mdatagen metrics builder (#6938)

v0.41.0

🛑 Breaking changes 🛑

  • None

🚀 New components 🚀

  • asapauthextension (#6627)
  • mongodbatlasreceiver (#6367)

🧰 Bug fixes 🧰

  • filestorageextension: fix panic when configured directory cannot be accessed (#6103)
  • hostmetricsreceiver: fix set of attributes for system.cpu.time metric (#6422)
  • k8sobserver: only record pod endpoints for running pods (#5878)
  • mongodbatlasreceiver: fix attributes fields in metadata.yaml (#6440)
  • prometheusexecreceiver: command line processing on Windows (#6145)
  • spanmetricsprocessor: fix exemplars support (#6140)
  • Remap arm64 to aarch64 on rpm/deb packages (#6635)

💡 Enhancements 💡

  • datadogexporter: do not use attribute localhost-like hostnames (#6477)
  • datadogexporter: retry per network call (#6412)
  • datadogexporter: take hostname into account for cache (#6223)
  • exporter/lokiexporter: adding a feature for loki exporter to encode JSON for log entry (#5846)
  • googlecloudspannerreceiver: added fallback to ADC for database connections. (#6629)
  • googlecloudspannerreceiver: added parsing only distinct items for sample lock request label. (#6514)
  • googlecloudspannerreceiver: added request tag label to metadata config for top query stats. (#6475)
  • googlecloudspannerreceiver: added sample lock requests label to the top lock stats metrics. (#6466)
  • googlecloudspannerreceiver: added transaction tag label to metadata config for top transaction stats. (#6433)
  • groupbyattrsprocessor: added support for metrics signal (#6248)
  • hostmetricsreceiver: ensure SchemaURL is set (#6482)
  • kubeletstatsreceiver: add support for read-only kubelet endpoint (#6488)
  • mysqlreceiver: enable native authentication (#6628)
  • mysqlreceiver: remove requirement for password on MySQL (#6479)
  • receiver/prometheusreceiver: do not add host.name to metrics from localhost/unspecified targets (#6476)
  • spanmetricsprocessor: add setStatus operation (#5886)
  • splunkhecexporter: remove duplication of host.name attribute (#6527)
  • tanzuobservabilityexporter: add consumer for sum metrics. (#6385)
  • Update log-collection library to v0.23.0 (#6593)

v0.40.0

🛑 Breaking changes 🛑

  • tencentcloudlogserviceexporter: change Endpoint to Region to simplify configuration (#6135)

🚀 New components 🚀

  • Add memcached receiver (#5839)

🧰 Bug fixes 🧰

  • Fix token passthrough for HEC (#5435)
  • datadogexporter: Fix missing resource attributes default mapping when resource_attributes_as_tags: false (#6359)
  • tanzuobservabilityexporter: Log and report missing metric values. (#5835)
  • mongodbatlasreceiver: Fix metrics metadata (#6395)

💡 Enhancements 💡

  • awsprometheusremotewrite exporter: Improve error message when failing to sign request
  • mongodbatlas: add metrics (#5921)
  • healthcheckextension: Add path option (#6111)
  • Set unprivileged user to container image (#6380)
  • k8sclusterreceiver: Add allocatable type of metrics (#6113)
  • observiqexporter: Allow Dialer timeout to be configured (#5906)
  • routingprocessor: remove broken debug log fields (#6373)
  • prometheusremotewriteexporter: Add exemplars support (#5578)
  • fluentforwardreceiver: Convert attributes with nil value to AttributeValueTypeEmpty (#6630)

v0.39.0

🛑 Breaking changes 🛑

  • httpdreceiver renamed to apachereceiver to match industry standards (#6207)
  • tencentcloudlogserviceexporter change Endpoint to Region to simplify configuration (#6135)

🚀 New components 🚀

  • Add postgresqlreceiver config and factory (#6153)
  • Add TencentCloud LogService exporter tencentcloudlogserviceexporter (#5722)
  • Restore jaegerthrifthttpexporter (#5666)
  • Add skywalkingexporter (#5690, #6114)

🧰 Bug fixes 🧰

  • datadogexporter: Improve cumulative metrics reset detection using StartTimestamp (#6120)
  • mysqlreceiver: Address issues in shutdown function (#6239)
  • tailsamplingprocessor: End go routines during shutdown (#5693)
  • googlecloudexporter: Update google cloud exporter to correctly close the metric exporter (#5990)
  • statsdreceiver: Fix the summary point calculation (#6155)
  • datadogexporter Correct default value for send_count_sum_metrics (#6130)

💡 Enhancements 💡

  • datadogexporter: Increase default timeout to 15 seconds (#6131)
  • googlecloudspannerreceiver: Added metrics cardinality handling for Google Cloud Spanner receiver (#5981, #6148, #6229)
  • mysqlreceiver: Mysql add support for different protocols (#6138)
  • bearertokenauthextension: Added support of Bearer Auth for HTTP Exporters (#5962)
  • awsxrayexporter: Fallback to rpc.method for segment operation when aws.operation missing (#6231)
  • healthcheckextension: Add new health check feature for collector pipeline (#5643)
  • datadogexporter: Always add current hostname (#5967)
  • k8sattributesprocessor: Add code to fetch all annotations and labels by specifying key regex (#5780)
  • datadogexporter: Do not rely on collector to resolve envvar when possible to resolve them (#6122)
  • datadogexporter: Add container tags to attributes package (#6086)
  • datadogexporter: Preserve original TraceID (#6158)
  • prometheusreceiver: Enhance prometheus receiver logger to determine errors, test real e2e usage (#5870)
  • awsxrayexporter: Added support for AWS AppRunner origin (#6141)

v0.38.0

🛑 Breaking changes 🛑

  • datadogexporter Make distributions the default histogram export option. (#5885)
  • redisreceiver Update Redis receiver's metric names. (#5837)
  • Remove scraperhelper from contrib, use the core version. (#5826)

🚀 New components 🚀

  • googlecloudspannerreceiver Added implementation of Google Cloud Spanner receiver. (#5727)
  • awsxrayproxy Wire up awsxrayproxy extension. (#5747)
  • awscontainerinsightreceiver Enable AWS Container Insight receiver. (#5960)

🧰 Bug fixes 🧰

  • statsdreceiver: fix start timestamp / temporality for counters. (#5714)
  • Fix security issue related to github.com/tidwall/gjson. (#5936)
  • datadogexporter Fix cumulative histogram handling in distributions mode (#5867)
  • datadogexporter Skip nil sketches (#5925)

💡 Enhancements 💡

  • Extend kafkareceiver configuration capabilities. (#5677)
  • Convert mongodbatlas receiver to use scraperhelper. (#5827)
  • Convert dockerstats receiver to use scraperhelper. (#5825)
  • Convert podman receiver to use scraperhelper. (#5822)
  • Convert redisreceiver to use scraperhelper. (#5796)
  • Convert kubeletstats receiver to use scraperhelper. (#5821)
  • googlecloudspannerreceiver Migrated Google Cloud Spanner receiver to scraper approach. (#5868)
  • datadogexporter Use a Consumer interface for decoupling from zorkian's package. (#5315)
  • mdatagen - Add support for extended metric descriptions (#5688)
  • signalfxexporter Log datapoints option. (#5689)
  • cumulativetodeltaprocessor: Update cumulative to delta. (#5772)
  • Update configuration default values in log receivers docs. (#5840)
  • fluentforwardreceiver: support more complex fluent-bit objects. (#5676)
  • datadogexporter Remove spammy logging. (#5856)
  • datadogexporter Remove obsolete report_buckets config. (#5858)
  • Improve performance of metric expression matcher. (#5864)
  • tanzuobservabilityexporter Introduce metricsConsumer and gaugeMetricConsumer. (#5426)
  • awsxrayexporter rpc.system has priority to determine aws namespace. (#5833)
  • tailsamplingprocessor Add support for composite sampling policy to the tailsampler. (#4958)
  • kafkaexporter Add support for AWS_MSK_IAM SASL Auth (#5763)
  • Refactor the client Authenticators for the new "ClientAuthenticator" interfaces (#5905)
  • mongodbatlasreceiver Add client wrapper for MongoDB Atlas support (#5386)
  • redisreceiver Update Redis config options (#5861)
  • routingprocessor: allow routing for all signals (#5869)
  • extension/observer/docker add ListAndWatch to observer (#5851)

v0.37.1

🧰 Bug fixes 🧰

  • Fixes a problem with v0.37.0 which contained dependencies on v0.36.0 components. They should have been updated to v0.37.0.

v0.37.0

🚀 New components 🚀

🛑 Breaking changes 🛑

  • Remove squash on configtls.TLSClientSetting for splunkhecexporter (#5541)
  • Remove squash on configtls.TLSClientSetting for elastic components (#5539)
  • Remove squash on configtls.TLSClientSetting for observiqexporter (#5540)
  • Remove squash on configtls.TLSClientSetting for AWS components (#5454)
  • Move k8sprocessor to k8sattributesprocessor.
  • Rename k8s_tagger configuration k8sattributes.
  • filelog receiver: use empty value for SeverityText field instead of "Undefined" (#5423)
  • Rename configparser.ConfigMap to config.Map
  • Rename pdata.AggregationTemporality* to pdata.MetricAggregationTemporality*
  • Remove deprecated batchpertrace package/module (#5380)

💡 Enhancements 💡

  • k8sattributes processor: add container metadata enrichment (#5467, #5572)
  • resourcedetection processor: Add an option to force using hostname instead of FQDN (#5064)
  • dockerstats receiver: Move docker client into new shared internal/docker (#4702)
  • spanmetrics processor:
    • Add exemplars to metrics (#5263)
    • Support resource attributes in metrics dimensions (#4624)
  • filter processor:
    • Add log filtering by regexp type filters (#5237)
    • Add record level log filtering (#5418)
  • dynatrace exporter: Handle non-gauge data types (#5056)
  • datadog exporter:
    • Add support for exporting histograms as sketches (#5082)
    • Scrub sensitive information from errors (#5575)
    • Add option to send instrumentation library metadata tags with metrics (#5431)
  • podman receiver: Add api_version, ssh_key, and ssh_passphrase config options (#5430)
  • signalfx exporter:
    • Add max_connections config option (#5432)
    • Add dimension name to log when value > 256 chars (#5258)
    • Discourage setting of endpoint path (#4851)
  • kubeletstats receiver: Convert to pdata instead of using OpenCensus (#5458)
  • tailsampling processor: Add invert_match config option to string_attribute policy (#4393)
  • awsemf exporter: Add a feature flag in UserAgent for AWS backend to monitor the adoptions (#5178)
  • splunkhec exporter: Handle explicitly NaN and Inf values (#5581)
  • hostmetrics receiver:
    • Collect more process states in processes scraper (#4856)
    • Add device label to paging scraper (#4854)
  • awskinesis exporter: Extend to allow for dynamic export types (#5440)

🧰 Bug fixes 🧰

  • datadog exporter:
    • Fix tags on summary and bucket metrics (#5416)
    • Fix cache key generation for cumulative metrics (#5417)
  • resourcedetection processor: Fix failure to start collector if at least one detector returns an error (#5242)
  • prometheus exporter: Do not record obsreport calls (#5438)
  • prometheus receiver: Metric type fixes to match Prometheus functionality (#4865)
  • sentry exporter: Fix sentry tracing (#4320)
  • statsd receiver: Set quantiles for metrics (#5647)

v0.36.0

🛑 Breaking changes 🛑

  • filter processor: The configs for logs filter processor have been changed to be consistent with the metrics filter processor. (#4895)
  • splunk_hec receiver:
    • source_key, sourcetype_key, host_key and index_key have now moved under hec_metadata_to_otel_attrs (#4726)
    • path field on splunkhecreceiver configuration is removed: We removed the path attribute as any request going to the Splunk HEC receiver port should be accepted, and added the raw_path field to explicitly map the path accepting raw HEC data. (#4951)
  • feat(dynatrace): tags is deprecated in favor of default_dimensions (#5055)

💡 Enhancements 💡

  • filter processor: Add ability to include logs based on resource attributes in addition to excluding logs based on resource attributes for strict matching. (#4895)
  • kubelet API: Add ability to create an empty CertPool when the system run environment is windows
  • JMX receiver: Allow JMX receiver logging level to be configured (#4898)
  • datadog exporter: Export histograms as in OpenMetrics Datadog check (#5065)
  • dockerstats receiver: Set Schema URL (#5239)
  • Rename memorylimiter -> memorylimiterprocessor (#5262)
  • awskinesis exporter: Refactor AWS kinesis exporter to be synchronous (#5248)

v0.35.0

🛑 Breaking changes 🛑

  • Rename configparser.Parser to configparser.ConfigMap (#5070)
  • Rename TelemetryCreateSettings -> TelemetrySettings (#5169)

💡 Enhancements 💡

  • chore: update influxdb exporter and receiver (#5058)
  • chore(dynatrace): use payload limit from api constants (#5077)
  • Add documentation for filelog's new force_flush_period parameter (#5066)
  • Reuse the gzip reader with a sync.Pool (#5145)
  • Add a trace observer when splunkhecreceiver is used for logs (#5063)
  • Remove usage of deprecated pdata.AttributeValueMapToMap (#5174)
  • Podman Stats Receiver: Receiver and Metrics implementation (#4577)

🧰 Bug fixes 🧰

  • Use staleness markers generated by prometheus, rather than making our own (#5062)
  • datadogexporter exporter: skip NaN and infinite values (#5053)

v0.34.0

🚀 New components 🚀

💡 Enhancements 💡

  • tailsampling processor: Add new policy probabilistic (#3876)

v0.33.0

🎉 OpenTelemetry Collector Contrib v0.33.0 (Beta) 🎉

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.

🚀 New components 🚀

💡 Enhancements 💡

  • Collector contrib has now full support for metrics proto v0.9.0.

v0.32.0

🎉 OpenTelemetry Collector Contrib v0.32.0 (Beta) 🎉

This release is marked as "bad" since the metrics pipelines will produce bad data.

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.

🛑 Breaking changes 🛑

  • splunk_hec receiver/exporter: com.splunk.source field is mapped to source field in Splunk instead of service.name (#4596)
  • redis receiver: Move interval runner package to internal/interval (#4600)
  • datadog exporter: Export summary count and sum as monotonic counts (#4605)

💡 Enhancements 💡

  • logzio exporter:
    • New implementation of an in-memory queue to store traces, data compression with gzip, and queue configuration options (#4395)
    • Make Hclog2ZapLogger struct and methods private for public go api review (#4431)
  • newrelic exporter (#4392):
    • Marked unsupported metric as permanent error
    • Force the interval to be valid even if 0
  • awsxray exporter: Add PHP stacktrace parsing support (#4454)
  • file_storage extension: Implementation of batch storage API (#4145)
  • datadog exporter:
    • Skip sum metrics with no aggregation temporality (#4597)
    • Export delta sums as counts (#4609)
  • elasticsearch exporter: Add dedot support (#4579)
  • signalfx exporter: Add process metric to translation rules (#4598)
  • splunk_hec exporter: Add profiling logs support (#4464)
  • awsemf exporter: Replace logGroup and logStream pattern with metric labels (#4466)

🧰 Bug fixes 🧰

  • awsxray exporter: Fix the origin on ECS/EKS/EB on EC2 cases (#4391)
  • splunk_hec exporter: Prevent re-sending logs that were successfully sent (#4467)
  • signalfx exporter: Prefix temporary metric translations (#4394)

v0.31.0

🎉 OpenTelemetry Collector Contrib v0.31.0 (Beta) 🎉

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.

🛑 Breaking changes 🛑

  • influxdb receiver: Removed metrics_schema config option (#4277)

💡 Enhancements 💡

  • Update to OTLP 0.8.0:
    • Remove use of IntHistogram (#4276)
    • Update exporters/receivers for NumberDataPoint
  • Remove use of deprecated pdata slice Resize() (#4203, #4208, #4209)
  • awsemf exporter: Added the option to have a user who is sending metrics from EKS Fargate Container Insights to reformat them to look the same as insights from ECS so that they can be ingested by CloudWatch (#4130)
  • k8scluster receiver: Support OpenShift cluster quota metrics (#4342)
  • newrelic exporter (#4278):
    • Requests are now retry-able via configuration option (defaults to retries enabled). Permanent errors are not retried.
    • The exporter monitoring metrics now include an untagged summary metric for ease of use.
    • Improved error logging to include URLs that fail to post messages to New Relic.
  • datadog exporter: Upscale trace stats when global sampling rate is set (#4213)

🧰 Bug fixes 🧰

  • statsd receiver: Add option to set Counter to be monotonic (#4154)
  • Fix internal/stanza severity mappings (#4315)
  • awsxray exporter: Fix the wrong AWS env resource setting (#4384)
  • newrelic exporter (#4278):
    • Configuration unmarshalling did not allow timeout value to be set to 0 in the endpoint specific section.
    • Request cancellation was not propagated via context into the http request.
    • The queued retry logger is set to a zap.Nop logger as intended.

v0.30.0

🎉 OpenTelemetry Collector Contrib v0.30.0 (Beta) 🎉

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.

🚀 New components 🚀

  • oauth2clientauth extension: ported from core (#3848)
  • metrics-generation processor: is now enabled and available (#4047)

🛑 Breaking changes 🛑

  • Removed jaegerthrifthttp exporter (#4089)

💡 Enhancements 💡

  • tailsampling processor:
    • Add new policy status_code (#3754)
    • Add new tail sampling processor policy: status_code (#3754)
  • awscontainerinsights receiver:
    • Integrate components and fix bugs for EKS Container Insights (#3846)
    • Add Cgroup to collect ECS instance metrics for container insights receiver #3875
  • spanmetrics processor: Support sub-millisecond latency buckets (#4091)
  • sentry exporter: Add exception event capture in sentry (#3854)

v0.29.0

🎉 OpenTelemetry Collector Contrib v0.29.0 (Beta) 🎉

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.

🛑 Breaking changes 🛑

  • redis receiver (#3808)
    • removed configuration service_name. Use resource processor or resource_attributes setting if using receivercreator
    • removed type label and set instrumentation library name to otelcol/redis as other receivers do

💡 Enhancements 💡

  • tailsampling processor:
    • Add new policy latency (#3750)
    • Add new policy status_code (#3754)
  • splunkhec exporter: Include trace_id and span_id if set (#3850)
  • newrelic exporter: Update instrumentation naming in accordance with otel spec (#3733)
  • sentry exporter: Added support for insecure connection with Sentry (#3446)
  • k8s processor:
    • Add namespace k8s tagger (#3384)
    • Add ignored pod names as config parameter (#3520)
  • awsemf exporter: Add support for TaskDefinitionFamily placeholder on log stream name (#3755)
  • loki exporter: Add resource attributes as Loki label (#3418)

🧰 Bug fixes 🧰

  • datadog exporter:
    • Ensure top level spans are computed (#3786)
    • Update env clobbering behavior (#3851)
  • awsxray exporter: Fixed filtered attribute translation (#3757)
  • splunkhec exporter: Include trace and span id if set in log record (#3850)

v0.28.0

🎉 OpenTelemetry Collector Contrib v0.28.0 (Beta) 🎉

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.

🚀 New components 🚀

🛑 Breaking changes 🛑

💡 Enhancements 💡

  • Enabled Dependabot for Github Actions (#3543)
  • Change obsreport helpers for receivers to use the new pattern created in Collector (#3439,#3443,#3449,#3504,#3521,#3548)
  • datadog exporter:
    • Add logging for unknown or unsupported metric types (#3421)
    • Add collector version tag to internal health metrics (#3394)
    • Remove sublayer stats calc and mutex (#3531)
    • Deduplicate hosts for which we send running metrics (#3539)
    • Add support for summary datatype (#3660)
    • Add datadog span operation name remapping config option (#3444)
    • Update error formatting for error spans that are not exceptions (#3701)
  • nginx receiver: Update the nginx metrics to more closely align with the conventions (#3420)
  • elasticsearch exporter: Init JSON encoding support (#3101)
  • jmx receiver:
    • Allow setting system properties (#3450)
    • Update tested JMX Metric Gatherer release (#3695)
  • Refactor components for the Client Authentication Extensions (#3507)
  • Remove redundant conversion calls (#3688)
  • storage extension: Add a Close method to Client interface (#3506)
  • splunkhec exporter: Add metric_type as key which maps to the type of the metric (#3696)
  • k8s processor: Add semantic conventions to k8s-tagger for pod metadata (#3544)
  • kubeletstats receiver: Refactor kubelet client to internal folder (#3698)
  • newrelic exporter (#3690):
    • Updates the log level from error to debug when New Relic rate limiting occurs
    • Updates the sanitized api key that is reported via metrics
  • filestorage extension: Add ability to specify name (#3703)
  • awsemf exporter: Store the initial value for cumulative metrics (#3425)
  • awskinesis exporter: Refactor to allow for extended types of encoding (#3655)
  • ecsobserver extension:
    • Add task definition, ec2, and service fetcher (#3503)
    • Add exporter to convert task to target (#3333)

🧰 Bug fixes 🧰

  • awsemf exporter: Remove delta adjustment from summaries by default (#3408)
  • alibabacloudlogservice exporter: Sanitize labels for metrics (#3454)
  • statsd receiver: Fix StatsD drop metrics tags when using summary as observer_type for timer/histogram (#3440)
  • awsxray exporter: Restore setting of Throttle for HTTP throttle response (#3685)
  • awsxray receiver: Fix quick start bug (#3653)
  • metricstransform processor: Check all data points for matching metric label values (#3435)

v0.27.0

🎉 OpenTelemetry Collector Contrib v0.27.0 (Beta) 🎉

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.

🚀 New components 🚀

💡 Enhancements 💡

  • splunkhec exporter:
    • Include the response in returned 400 errors (#3338)
    • Map summary metrics to Splunk HEC metrics (#3344)
    • Add HEC telemetry (#3260)
  • newrelic exporter: Include dropped attributes and events counts (#3187)
  • datadog exporter:
    • Add Fargate task ARN to container tags (#3326)
    • Improve mappings for span kind dd span type (#3368)
  • signalfx exporter: Add info log for host metadata properties update (#3343)
  • awsprometheusremotewrite exporter: Add SDK and system information to User-Agent header (#3317)
  • metricstransform processor: Add filtering capabilities matching metric label values for applying changes (#3201)
  • groupbytrace processor: Added workers for queue processing (#2902)
  • resourcedetection processor: Add docker detector (#2775)
  • tailsampling processor: Support regex on span attribute filtering (#3335)

🧰 Bug fixes 🧰

  • datadog exporter:
    • Update Datadog attributes to tags mapping (#3292)
    • Consistent hostname and default metrics behavior (#3286)
  • signalfx exporter: Handle character limits on metric names and dimensions (#3328)
  • newrelic exporter: Fix timestamp value for cumulative metrics (#3406)

v0.26.0

🎉 OpenTelemetry Collector Contrib v0.26.0 (Beta) 🎉

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.

🚀 New components 🚀

  • influxdb exporter to support sending tracing, metrics, and logging data to InfluxDB

🛑 Breaking changes 🛑

  • signalfx exporter (#3207):
    • Additional metrics excluded by default by signalfx exporter
      • system.disk.io_time
      • system.disk.operation_time
      • system.disk.weighted_io_time
      • system.network.connections
      • system.processes.count
      • system.processes.created

💡 Enhancements 💡

  • Add default config and systemd environment file support for DEB/RPM packages (#3123)
  • Log errors on receiver start/stop failures (#3208)
  • newrelic exporter: Update API key detection logic (#3212)
  • splunkhec exporter:
    • Mark permanent errors to avoid futile retries (#3253)
    • Add TLS certs verification (#3204)
  • datadog exporter:
    • Add env and tag name normalization to trace payloads (#3200)
    • add ignore_resources configuration option (#3245)
  • jmx receiver: Update for latest snapshot and header support (#3283)
  • awsxray exporter: Added support for stack trace translation for .NET language (#3280)
  • statsd receiver: Add timing/histogram for statsD receiver as OTLP summary (#3261)

🧰 Bug fixes 🧰

  • awsprometheusremotewrite exporter:
    • Remove sending_queue (#3186)
    • Use the correct default for aws_auth.service (#3161)
    • Identify the Amazon Prometheus region from the endpoint (#3210)
    • Don't panic in case session can't be constructed (#3221)
  • datadog exporter: Add max tag length (#3185)
  • sapm exporter: Fix crash when passing the signalfx access token (#3294)
  • newrelic exporter: Update error conditions (#3322)

v0.25.0

🎉 OpenTelemetry Collector Contrib v0.25.0 (Beta) 🎉

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.

🚀 New components 🚀

  • kafkametricsreceiver new receiver component for collecting metrics about a kafka cluster - primarily lag and offset. configuration instructions
  • file_storage extension to read and write data to the local file system (#3087)

🛑 Breaking changes 🛑

  • newrelic exporter (#3091):
    • Removal of common attributes (use opentelemetry collector resource processor to add attributes)
    • Drop support for cumulative metrics being sent to New Relic via a collector

💡 Enhancements 💡

  • Update opentelemetry-log-collection to v0.17.0 for log receivers (#3017)
  • datadog exporter:
    • Add peer.service priority instead of service.name (#2817)
    • Improve support of semantic conventions for K8s, Azure and ECS (#2623)
  • Improve and batch logs translation for stanza (#2892)
  • statsd receiver: Add timing/histogram as OTLP gauge (#2973)
  • honeycomb exporter: Add Retry and Queue settings (#2714)
  • resourcedetection processor:
    • Add AKS resource detector (#3035)
    • Use conventions package constants for ECS detector (#3171)
  • sumologic exporter: Add graphite format (#2695)
  • Add trace attributes to the log entry for stanza (#3018)
  • splunk_hec exporter: Send log record name as part of the HEC log event (#3119)
  • newrelic exporter (#3091):
    • Add support for logs
    • Performance improvements
    • Optimizations to the New Relic payload to reduce payload size
    • Metrics generated for monitoring the exporter
    • Insert Key vs License keys are auto-detected in some cases
    • Collector version information is properly extracted via the application start info parameters

🧰 Bug fixes 🧰

  • splunk_hec exporter: Fix sending log payload with missing the GZIP footer (#3032)
  • awsxray exporter: Remove propagation of error on shutdown (#2999)
  • resourcedetection processor:
    • Correctly report DRAGONFLYBSD value (#3100)
    • Fallback to os.Hostname when FQDN is not available (#3099)
  • httpforwarder extension: Do not report ErrServerClosed when shutting down the service (#3173)
  • collectd receiver: Do not report ErrServerClosed when shutting down the service (#3178)

v0.24.0

🎉 OpenTelemetry Collector Contrib v0.24.0 (Beta) 🎉

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.

🚀 New components 🚀

  • fluentbit extension and fluentforward receiver moved from opentelemetry-collector

💡 Enhancements 💡

  • Check NO_WINDOWS_SERVICE environment variable to force interactive mode on Windows (#2819)
  • resourcedetection processor:
    • Add task revision to ECS resource detector (#2814)
    • Add GKE detector (#2821)
    • Add Amazon EKS detector (#2820)
    • Add VMScaleSetName field to Azure detector (#2890)
  • awsemf exporter:
    • Add parse_json_encoded_attr_values config option to decode json-encoded strings in attribute values (#2827)
    • Add output_destination config option to support AWS Lambda (#2720)
  • googlecloud exporter: Handle cloud.availability_zone semantic convention (#2893)
  • newrelic exporter: Add instrumentation.provider to default attributes (#2900)
  • Set unprivileged user to container image (#2925)
  • splunkhec exporter: Add max_content_length_logs config option to send log data in payloads less than max content length (#2524)
  • k8scluster and kubeletstats receiver: Replace package constants in favor of constants from conventions in core (#2996)

🧰 Bug fixes 🧰

  • spanmetrics processor:
    • Rename calls metric to calls_total and set IsMonotonic to true (#2837)
    • Validate duplicate dimensions at start (#2844)
  • awsemf exporter: Calculate delta instead of rate for cumulative metrics (#2512)
  • signalfx exporter:
    • Remove more unnecessary translation rules (#2889)
    • Implement summary type (#2998)
  • awsxray exporter: Remove translation to HTTP status from OC status (#2978)
  • awsprometheusremotewrite exporter: Close HTTP body after RoundTrip (#2955)
  • splunkhec exporter: Add ResourceAttributes to Splunk Event (#2843)

v0.23.0

🎉 OpenTelemetry Collector Contrib v0.23.0 (Beta) 🎉

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.

🚀 New components 🚀

  • groupbyattrs processor to group the records by provided attributes
  • dotnetdiagnostics receiver to read metrics from .NET processes

🛑 Breaking changes 🛑

  • stackdriver exporter marked as deprecated and renamed to googlecloud
  • Change the rule expression in receiver creator for matching endpoints types from type.port, type.hostport and type.pod to type == "port", type == "hostport" and type == "pod" (#2661)

💡 Enhancements 💡

  • loadbalancing exporter: Add support for logs (#2470)
  • sumologic exporter: Add carbon formatter (#2562)
  • awsecscontainermetrics receiver: Add new metric for stopped container (#2383)
  • awsemf exporter:
    • Send EMF logs in batches (#2572)
    • Add prometheus type field for CloudWatch compatibility (#2689)
  • signalfx exporter:
    • Add resource attributes to events (#2631)
    • Add translation rule to drop dimensions (#2660)
    • Remove temporary host translation workaround (#2652)
    • Remove unnecessary default translation rules (#2672)
    • Update exclude_metrics option so that the default exclude rules can be overridden by setting the option to [] (#2737)
  • awsprometheusremotewrite exporter: Add support for given IAM roles (#2675)
  • statsd receiver: Change to use OpenTelemetry type instead of OpenCensus type (#2733)
  • resourcedetection processor: Add missing entries for cloud.infrastructure_service (#2777)

🧰 Bug fixes 🧰

  • dynatrace exporter: Serialize each datapoint into separate line (#2618)
  • splunkhec exporter: Retain all otel attributes (#2712)
  • newrelic exporter: Fix default metric URL (#2739)
  • googlecloud exporter: Add host.name label if hostname is present in node (#2711)

v0.22.0

🎉 OpenTelemetry Collector Contrib v0.22.0 (Beta) 🎉

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.

🚀 New components 🚀

💡 Enhancements 💡

  • dynatrace exporter: Send metrics to Dynatrace in chunks of 1000 (#2468)
  • k8s processor: Add ability to associate metadata tags using pod UID rather than just IP (#2199)
  • signalfx exporter:
    • Add statusCode to logging field on dimension client (#2459)
    • Add translation rules for cpu.utilization_per_core (#2540)
    • Updates to metadata handling (#2531)
    • Calculate extra network I/O metrics (#2553)
    • Calculate extra disk I/O metrics (#2557)
  • statsd receiver: Add metric type label and enable_metric_type option (#2466)
  • sumologic exporter: Add support for carbon2 format (#2562)
  • resourcedetection processor: Add Azure detector (#2372)
  • k8scluster receiver: Use OTel conventions for metadata (#2530)
  • newrelic exporter: Multi-tenant support for sending trace data and performance enhancements (#2481)
  • stackdriver exporter: Enable retry_on_failure and sending_queue options (#2613)
  • Use standard way to convert from time.Time to proto Timestamp (#2548)

🧰 Bug fixes 🧰

  • signalfx exporter:
    • Fix calculation of network.total metric (#2551)
    • Correctly convert dimensions on metadata updates (#2552)
  • awsxray exporter and receiver: Fix the type of content_length (#2539)
  • resourcedetection processor: Use values in accordance to semantic conventions for AWS (#2556)
  • awsemf exporter: Fix concurrency issue (#2571)

v0.21.0

🎉 OpenTelemetry Collector Contrib v0.21.0 (Beta) 🎉

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.

🚀 New components 🚀

  • loki exporter to export data via HTTP to Loki

🛑 Breaking changes 🛑

  • signalfx exporter: Allow periods to be sent in dimension keys (#2456). Existing users who do not want to change this functionality can set nonalphanumeric_dimension_chars to _-

💡 Enhancements 💡

  • awsemf exporter:
    • Support unit customization before sending logs to AWS CloudWatch (#2318)
    • Group exported metrics by labels (#2317)
  • datadog exporter: Add basic span events support (#2338)
  • alibabacloudlogservice exporter: Support new metrics interface (#2280)
  • sumologic exporter:
    • Enable metrics pipeline (#2117)
    • Add support for all types of log body (#2380)
  • signalfx exporter: Add nonalphanumeric_dimension_chars config option (#2442)

🧰 Bug fixes 🧰

  • resourcedetection processor: Fix resource attribute environment variable (#2378)
  • k8scluster receiver: Fix nil pointer bug (#2450)

v0.20.0

🎉 OpenTelemetry Collector Contrib v0.20.0 (Beta) 🎉

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.

🚀 New components 🚀

  • spanmetrics processor to aggregate Request, Error and Duration (R.E.D) metrics from span data
  • awsxray receiver to accept spans in the X-Ray Segment format
  • groupbyattrs processor to group the records by provided attributes

🛑 Breaking changes 🛑

  • Rename kinesis exporter to awskinesis (#2234)
  • signalfx exporter: Remove send_compatible_metrics option, use translation_rules instead (#2267)
  • datadog exporter: Remove default prefix from user metrics (#2308)

💡 Enhancements 💡

  • signalfx exporter: Add k8s metrics to default excludes (#2167)
  • stackdriver exporter: Reduce QPS (#2191)
  • datadog exporter:
    • Translate otel exceptions to DataDog errors (#2195)
    • Use resource attributes for metadata and generated metrics (#2023)
  • sapm exporter: Enable queuing by default (#1224)
  • dynatrace exporter: Allow underscores anywhere in metric or dimension names (#2219)
  • awsecscontainermetrics receiver: Handle stopped container's metadata (#2229)
  • awsemf exporter: Enhance metrics batching in AWS EMF logs (#2271)
  • f5cloud exporter: Add User-Agent header with version to requests (#2292)

🧰 Bug fixes 🧰

  • signalfx exporter: Reinstate network/filesystem translation rules (#2171)

v0.19.0

🎉 OpenTelemetry Collector Contrib v0.19.0 (Beta) 🎉

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.

🚀 New components 🚀

  • f5cloud exporter to export metric, trace, and log data to F5 Cloud
  • jmx receiver to report metrics from a target MBean server in conjunction with the JMX Metric Gatherer

🛑 Breaking changes 🛑

  • signalfx exporter: The exclude_metrics option now takes slice of metric filters instead of just metric names (slice of strings) (#1951)

💡 Enhancements 💡

  • datadog exporter: Sanitize datadog service names (#1982)
  • awsecscontainermetrics receiver: Add more metadata (#2011)
  • azuremonitor exporter: Favor RPC over HTTP spans (#2006)
  • awsemf exporter: Always use float64 as calculated rate (#2019)
  • splunkhec receiver: Make the HEC receiver path configurable, and use /* by default (#2137)
  • signalfx exporter:
    • Drop non-default metrics and add include_metrics option to override (#2145, #2146, #2162)
    • Rename system.network.dropped_packets metric to system.network.dropped (#2160)
    • Do not filter cloud attributes from dimensions (#2020)
  • redis receiver: Migrate to pdata metrics #1889

🧰 Bug fixes 🧰

  • datadog exporter: Ensure that version tag is added to trace stats (#2010)
  • loadbalancing exporter: Rolling update of collector can stop the periodical check of DNS updates (#1798)
  • awsecscontainermetrics receiver: Change the type of exit_code from string to int and deal with the situation when there is no data (#2147)
  • groupbytrace processor: Make onTraceReleased asynchronous to fix processor overload (#1808)
  • Handle cases where the time field of Splunk HEC events is encoded as a String (#2159)

v0.18.0

🎉 OpenTelemetry Collector Contrib v0.18.0 (Beta) 🎉

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.

🚀 New components 🚀

  • sumologic exporter to send logs and metrics data to Sumo Logic
  • dynatrace exporter to send metrics to Dynatrace

💡 Enhancements 💡

  • datadog exporter:
    • Add resource attributes to tags conversion feature (#1782)
    • Add Kubernetes conventions for hostnames (#1919)
    • Add container tags to datadog export for container infra metrics in service view (#1895)
    • Update resource naming and span naming (#1861)
    • Add environment variables support for config options (#1897)
  • awsxray exporter: Add parsing of JavaScript stack traces (#1888)
  • elastic exporter: Translate exception span events (#1858)
  • signalfx exporter: Add translation rules to aggregate per core CPU metrics in default translations (#1841)
  • resourcedetection processor: Gather tags associated with the EC2 instance and add them as resource attributes (#1899)
  • simpleprometheus receiver: Add support for passing params to the prometheus scrape config (#1949)
  • azuremonitor exporter: Implement Span status code specification changes - gRPC (#1960)
  • metricstransform processor: Add grouping option ($1887)
  • alibabacloudlogservice exporter: Use producer to send data to improve performance (#1981)

🧰 Bug fixes 🧰

  • datadog exporter: Handle monotonic metrics client-side (#1805)
  • awsxray exporter: Log error when translating span (#1809)

v0.17.0

🎉 OpenTelemetry Collector Contrib v0.17.0 (Beta) 🎉

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.

💡 Enhancements 💡

  • awsemf exporter: Add collector version to EMF exporter user agent (#1778)
  • signalfx exporter: Add configuration for trace correlation (#1795)
  • statsd receiver: Add support for metric aggregation (#1670)
  • datadog exporter: Improve logging of hostname detection (#1796)

🧰 Bug fixes 🧰

  • resourcedetection processor: Fix ecs detector to not use the default golang logger (#1745)
  • signalfx receiver: Return 200 when receiver succeed (#1785)
  • datadog exporter: Use a singleton for sublayer calculation (#1759)
  • awsxray and awsemf exporters: Change the User-Agent content order (#1791)

v0.16.0

🎉 OpenTelemetry Collector Contrib v0.16.0 (Beta) 🎉

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.

🛑 Breaking changes 🛑

  • honeycomb exporter: Update to use internal data format (#1689)

💡 Enhancements 💡

  • newrelic exporter: Add support for span events (#1643)
  • awsemf exporter:
    • Add placeholder support in log_group_name and log_stream_name config (#1623, #1661)
    • Add label matching filtering rule (#1619)
  • 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)

🧰 Bug fixes 🧰

  • 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)

v0.15.0

🎉 OpenTelemetry Collector Contrib v0.15.0 (Beta) 🎉

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.

🚀 New components 🚀

  • zookeeper receiver: Collects metrics from a Zookeeper instance using the mntr command
  • loadbalacing exporter: Consistently exports spans belonging to the same trace to the same backend
  • windowsperfcounters receiver: Captures the configured system, application, or custom performance counter data from the Windows registry using the PDH interface
  • awsprometheusremotewrite 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

💡 Enhancements 💡

  • awsemf exporter:
    • Add metric_declarations config option for metric filtering and dimensions (#1503)
    • Add SummaryDataType and remove Min/Max from Histogram (#1584)
  • 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)
  • awsecscontainermetrics receiver: Extract cluster name from ARN (#1626)
  • elastic exporter: Improve handling of span status if the status code is unset (#1591)

🧰 Bug fixes 🧰

  • awsemf exporter: Add check for unhandled metric data types (#1493)
  • groupbytrace processor: Make buffered channel to avoid goroutines leak (#1505)
  • stackdriver exporter: Set options.UserAgent so that the OpenCensus exporter does not override the UA ($1620)

v0.14.0

🎉 OpenTelemetry Collector Contrib v0.14.0 (Beta) 🎉

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.

🚀 New components 🚀

  • datadog exporter to send metric and trace data to Datadog (#1352)
  • tailsampling processor moved from core to contrib (#1383)

🛑 Breaking changes 🛑

  • jmxmetricsextension migrated to jmxreceiver (#1182, #1357)
  • Move signalfx correlation code out of sapm to signalfxcorrelation exporter (#1376)
  • Move Splunk specific utils outside of common (#1306)
  • stackdriver exporter:
  • datadog exporter: Remove option to change the namespace prefix (#1483)

💡 Enhancements 💡

  • 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)

🧰 Bug fixes 🧰

  • 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)

v0.13.0

🎉 OpenTelemetry Collector Contrib v0.13.0 (Beta) 🎉

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.

💡 Enhancements 💡

  • 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: Add span.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: Report CpuUtilized metric in percentage (#1283)
  • awsemf exporter: Only calculate metric rate for cumulative counter and avoid SingleDimensionRollup for metrics with only one dimension (#1280)

🧰 Bug fixes 🧰

  • 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)

v0.12.0

🎉 OpenTelemetry Collector Contrib v0.12.0 (Beta) 🎉

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.

🚀 New components 🚀

  • 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)

💡 Enhancements 💡

  • 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)

🧰 Bug fixes 🧰

  • awsxray exporter: Don't set origin to EC2 when not on AWS (#1115)

v0.11.0

🎉 OpenTelemetry Collector Contrib v0.11.0 (Beta) 🎉

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.

🚀 New components 🚀

  • add dockerstats receiver as top level component (#1081)
  • add tracegen utility (#956)

💡 Enhancements 💡

  • stackdriver exporter: Allow overriding client options via config (#1010)
  • k8scluster receiver: Ensure informer caches are synced before initial data sync (#842)
  • elastic exporter: Translate deployment.environment resource attribute to Elastic APM's semantically equivalent service.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 and disk_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)

v0.10.0

🎉 OpenTelemetry Collector Contrib v0.10.0 (Beta) 🎉

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.

🚀 New components 🚀

  • 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)

💡 Enhancements 💡

  • 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)

🧰 Bug fixes 🧰

  • 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)

v0.9.0

🎉 OpenTelemetry Collector Contrib v0.9.0 (Beta) 🎉

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.

🛑 Breaking changes 🛑

  • Remove deprecated lightstep exporter (#828)

🚀 New components 🚀

  • statsd receiver for ingesting StatsD messages (#566)

💡 Enhancements 💡

  • 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)

🧰 Bug fixes 🧰

  • 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)

v0.8.0

🎉 OpenTelemetry Collector Contrib v0.8.0 (Beta) 🎉

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.

🚀 New components 🚀

  • Receivers
    • prometheusexec subprocess manager (##499)

💡 Enhancements 💡

  • 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 to k8s.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 to k8s.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: Change container.spec.name label to k8s.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)

🧰 Bug fixes 🧰

  • 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)

🚀 New components 🚀

  • Skeleton for AWS ECS container metrics receiver (#463)
  • prometheus_exec receiver (#655)

v0.7.0

🎉 OpenTelemetry Collector Contrib v0.7.0 (Beta) 🎉

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.

🛑 Breaking changes 🛑

  • awsxray receiver updated to support udp: tcp_endpoint config option renamed to endpoint (#497)
  • TLS config changed for sapmreceiver (#488) and signalfxreceiver receivers (#488)

🚀 New components 🚀

  • 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)

💡 Enhancements 💡

  • 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 in kubeletstats receiver to add extra labels (#569)
  • Add metric translation options to signalfx exporter (#477, #501, #571, #573)

🧰 Bug fixes 🧰

  • azuremonitor exporter: Mark spanToEnvelope errors as permanent (#500)

v0.6.0

🎉 OpenTelemetry Collector Contrib v0.6.0 (Beta) 🎉

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.

🛑 Breaking changes 🛑

  • Removed jaegarlegacy (#397) and zipkinscribe receivers (#410)
  • kubeletstats receiver: Renamed k8s.pod.namespace pod label to k8s.namespace.name and k8s.container.name container label to container.name

🚀 New components 🚀

  • Processors
    • metricstransform renames/aggregates within individual metrics (#376) and allow changing the data type between int and float (#402)

💡 Enhancements 💡

  • awsxray exporter: Use peer.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)

🧰 Bug fixes 🧰

  • 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)

v0.5.0

Released 01-07-2020

🎉 OpenTelemetry Collector Contrib v0.5.0 (Beta) 🎉

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.

🚀 New components 🚀

  • Processors
    • resourcedetection to automatically detect the resource based on the configured set of detectors (#309)

💡 Enhancements 💡

  • 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)

🧰 Bug fixes 🧰

  • honeycomb exporter: Remove shared use of libhoney from goroutines (#305)

v0.4.0

Released 17-06-2020

🎉 OpenTelemetry Collector Contrib v0.4.0 (Beta) 🎉

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.

🛑 Breaking changes 🛑

  • signalfx exporter url parameter changed to ingest_url (no impact if only using realm setting)

🚀 New components 🚀

  • 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)

💡 Enhancements 💡

  • 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 just ip (#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)

🧰 Bug fixes 🧰

  • 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)

🧩 Components 🧩

Receivers

Traces Metrics
Jaeger Legacy Carbon
SAPM (SignalFx APM) Collectd
Zipkin Scribe K8s Cluster
Redis
SignalFx
Simple Prometheus
Wavefront

Processors

  • K8s

Exporters

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)

Extensions

  • Observer
    • K8s

v0.3.0 Beta

Released 2020-03-30

Breaking changes

  • Make prometheus receiver config loading strict. #697 Prometheus receiver will now fail fast if the config contains unused keys in it.

Changes and fixes

  • 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

Features

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

v0.2.8

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.

v0.2.7

Released 2020-03-17

Self-Observability

  • 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.

Processors

  • Fixed passthrough mode k8sprocessor.
  • Added HASH action to attribute processor.

Receivers and Exporters

  • 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.

v0.2.6 Alpha

Released 2020-02-18

Self-Observability

  • 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.

Processors

  • Extended Span processor to have include/exclude span logic.
  • Ability to choose strict or regexp matching for include/exclude filters.

Receivers and Exporters

  • Added Carbon receiver and exporter.
  • Added Wavefront receiver.

v0.0.5 Alpha

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.

v0.0.1 Alpha

Released 2020-01-11

First release of OpenTelemetry Collector Contrib.