Skip to content

Latest commit

 

History

History
64 lines (50 loc) · 2.74 KB

8.6.asciidoc

File metadata and controls

64 lines (50 loc) · 2.74 KB

APM version 8.6

APM version 8.6.2

No significant changes.

APM version 8.6.1

Bug fixes

  • Fix panic when memory limits are missing in cgroupV1 {pull}10098[10098]

  • Fix internal traffic filter handling on Elastic Cloud when managed by Elastic Agent (Integrations Server)

APM version 8.6.0

Breaking Changes

  • apm-server.decoder.* stack monitoring metrics are no longer recorded {pull}9210[9210]

  • ecs.version is no longer added to document _source; it is added as a constant_keyword field {pull}9208[9208]

  • context.http.response.*_size fields now enforce integer values {pull}9429[9429]

  • observer.id and observer.ephemeral_id are no longer added to APM documents {pull}9412[9412]

  • timeseries.instance has been removed from transaction metrics docs; it was never used {pull}9565[9565]

Deprecations

  • apm-server apikey commands have been deprecated. API Keys should be managed through Kibana or the Elasticsearch REST API {pull}9446[9446]

Bug fixes

  • Change ownership of apm-server.yml and data/ to apm-server user in Docker image {pull}9604[9604]

Intake API Changes

  • experimental:[] Extend logs v2 intake API to support ECS logging fields {pull}9349[9349]

  • experimental:[] Add support for string timestamp format(2006-01-02T15:04:05.999-0700) {pull}9376[9376]

  • context.http.response.*_size fields have been changed from floating-point to integer types {pull}9429[9429]

Added

  • Service metrics now support global labels {pull}9248[9248]

  • OpenTelemetry GRPC Spans from the Javascript API/SDK/Instrumentations are now correctly transformed into transactions with type=request {pull}9308[9308]

  • Improve Elasticsearch output performance, particularly when compression is enabled (default) {pull}9318[9318]

  • Java attacher support for macOS {pull}9413[9413]

  • Improve Elasticsearch output performance in instances with more than 6 cores {pull}9393[9393]

  • Tune internal server configuration based on available cgroup or machine RAM, improving resource utilization and performance {pull}9358[9358]

  • Disallow auto-scaling of active indexers when Elasticsearch 429 response rate exceeds 1% of total requests issued {pull}9463[9463]

  • We now record transaction.representative_count and span.representative_count — the inverse sample rate {pull}9458[9458]