## Validate using https://www.otelbin.io/#distro=adot%7E&distroVersion=v0.43.1%7E . ## ## Original copyright is Apache 2.0 to AWS (aws-observability on GitHub). ## https://github.com/aws-observability/aws-otel-collector/blob/14833d4af543da709c77cf9dc6827351dbd529b1/config/ecs/container-insights/otel-task-metrics-config.yaml ## ## This is a modified version of `./etc/ecs/container-insights/otel-task-metrics-config.yaml` ## where we add `resource_to_telemetry_conversion` so the otel resource `service.name` ## can be added to the metric. This is useful because you can filter in metrics based on service, ## instead of grouping e.g. all services' memory usage under the same metric. ## ## Search for `# THIS IS WHAT LIFLIG ADDED` to find our modifications. ## ## ## @see https://aws-otel.github.io/docs/setup/ecs/config-through-ssm extensions: health_check: ## These ports are also set in `taskDefinition.defaultContainer` as ## `OTEL_EXPORTER_OTLP_ENDPOINT` and so on. ## Changes must happen in both places. receivers: otlp: protocols: grpc: endpoint: 0.0.0.0:4317 http: endpoint: 0.0.0.0:4318 awsxray: endpoint: 0.0.0.0:2000 transport: udp awsecscontainermetrics: processors: batch/traces: timeout: 1s send_batch_size: 50 batch/metrics: timeout: 60s filter: metrics: include: match_type: strict metric_names: - ecs.task.memory.reserved - ecs.task.memory.utilized - ecs.task.cpu.reserved - ecs.task.cpu.utilized - ecs.task.network.rate.rx - ecs.task.network.rate.tx - ecs.task.storage.read_bytes - ecs.task.storage.write_bytes - container.duration metricstransform: transforms: - include: ecs.task.memory.utilized action: update new_name: MemoryUtilized - include: ecs.task.memory.reserved action: update new_name: MemoryReserved - include: ecs.task.cpu.utilized action: update new_name: CpuUtilized - include: ecs.task.cpu.reserved action: update new_name: CpuReserved - include: ecs.task.network.rate.rx action: update new_name: NetworkRxBytes - include: ecs.task.network.rate.tx action: update new_name: NetworkTxBytes - include: ecs.task.storage.read_bytes action: update new_name: StorageReadBytes - include: ecs.task.storage.write_bytes action: update new_name: StorageWriteBytes ## THIS IS WHAT LIFLIG ADDED filter/application: error_mode: ignore metrics: metric: - 'resource.attributes["http.target"] == "/health"' - 'resource.attributes["net.peer.name"] == "169.254.170.2"' # EC2 Metadata Endpoint filter/traces: error_mode: ignore traces: span: - 'attributes["http.route"] == "health"' - 'attributes["http.url"] == "/health"' - 'name == "GET health"' attributes/application: actions: # Http adds lots of high cardinality, which xray creates 1 metric per. Expensive. - key: http.url action: delete - key: http.target action: delete - key: http.client_ip action: delete - key: http.response_content_length action: delete - key: http.request_content_length action: delete - key: net.host.name action: delete - key: net.sock.peer.addr action: delete - key: net.sock.peer.port action: delete - key: net.sock.host.addr action: delete - key: net.sock.host.port action: delete - key: user_agent.original action: delete resource/application: attributes: - key: cloud.provider action: delete - key: host.arch action: delete - key: aws.ecs.container.image.id action: delete - key: aws.ecs.task.arn action: delete - key: aws.log.stream.names action: delete - key: aws.log.stream.arns action: delete - key: cloud.platform action: delete - key: container.name action: delete - key: process.executable.path action: delete - key: process.runtime.version action: delete - key: telemetry.auto.version action: delete - key: telemetry.sdk.name action: delete - key: container.id action: delete - key: container.image.tag action: delete - key: process.runtime.name action: delete - key: service.namespace action: delete - key: telemetry.sdk.version action: delete - key: aws.ecs.task.family action: delete - key: aws.ecs.task.revision action: delete - key: aws.log.group.arns action: delete - key: container.image.name action: delete - key: os.description action: delete - key: os.type action: delete - key: process.runtime.description action: delete - key: version action: delete - key: aws.ecs.launchtype action: delete - key: aws.log.group.names action: delete - key: host.name action: delete - key: process.pid action: delete - key: telemetry.sdk.language action: delete - key: aws.ecs.container.arn action: delete - key: service.version action: delete - key: process.command_args action: delete - key: service action: delete ## END OF LIFLIG CHANGES resource: attributes: - key: ClusterName from_attribute: aws.ecs.cluster.name action: insert - key: aws.ecs.cluster.name action: delete - key: ServiceName from_attribute: aws.ecs.service.name action: insert - key: aws.ecs.service.name action: delete ## THIS IS WHAT LIFLIG ADDED (bunch of sporadic removals) - key: aws.ecs.task.id action: delete - key: TaskDefinitionFamily from_attribute: aws.ecs.task.family action: insert - key: aws.ecs.task.family action: delete - key: aws.ecs.task.arn action: delete - key: aws.ecs.docker.name action: delete - key: aws.ecs.task.version action: delete - key: aws.ecs.task.pull_started_at action: delete - key: aws.ecs.task.pull_stopped_at action: delete - key: AvailabilityZone from_attribute: cloud.zone action: insert - key: cloud.zone action: delete - key: aws.ecs.task.launch_type action: delete - key: Region from_attribute: cloud.region action: insert - key: cloud.region action: delete - key: AccountId from_attribute: cloud.account.id action: insert - key: cloud.account.id action: delete - key: container.id action: delete - key: container.name action: delete - key: container.image.name action: delete - key: aws.ecs.container.image.id action: delete - key: aws.ecs.container.exit_code action: delete - key: aws.ecs.container.created_at action: delete - key: aws.ecs.container.started_at action: delete - key: aws.ecs.container.finished_at action: delete - key: container.image.tag action: delete ## END OF LIFLIG CHANGES exporters: awsxray: ## THIS IS WHAT LIFLIG ADDED indexed_attributes: ["otel.resource.service.name"] # Max 50 attributes ## END OF LIFLIG CHANGES awsemf/application: namespace: ECS/AWSOTel/Application log_group_name: '/aws/ecs/application/metrics' ## THIS IS WHAT LIFLIG ADDED ## Config docs at https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/main/exporter/awsemfexporter/README.md ## log retention is days to keep the emf metric logs. log_retention: 365 resource_to_telemetry_conversion: enabled: true dimension_rollup_option: NoDimensionRollup ## No rollup means we keep all the dimensions on the metric, instead of merging similar metrics to one in CloudWatch ## END OF LIFLIG CHANGES awsemf/performance: namespace: ECS/ContainerInsights log_group_name: '/aws/ecs/containerinsights/{ClusterName}/performance' log_stream_name: '{TaskId}' resource_to_telemetry_conversion: enabled: true dimension_rollup_option: NoDimensionRollup metric_declarations: - dimensions: [ [ ClusterName ], [ ClusterName, TaskDefinitionFamily ] ] metric_name_selectors: - MemoryUtilized - MemoryReserved - CpuUtilized - CpuReserved - NetworkRxBytes - NetworkTxBytes - StorageReadBytes - StorageWriteBytes - metric_name_selectors: [container.*] service: pipelines: traces: receivers: [otlp,awsxray] ## THIS IS WHAT LIFLIG ADDED processors: [filter/traces, batch/traces] exporters: [awsxray] metrics/application: receivers: [otlp] processors: [filter/application, resource/application, attributes/application, batch/metrics] exporters: [awsemf/application] ## END OF LIFLIG CHANGES metrics/performance: receivers: [awsecscontainermetrics ] processors: [filter, metricstransform, resource] exporters: [ awsemf/performance ] extensions: [health_check]