APM version 8.7

edit
APM version 8.7.2
edit

View commits

Bug fixesedit
APM version 8.7.1
edit

View commits

Bug fixesedit
  • Fix indexing failures for metric, error and log documents when agent.activation_method is set 10552
  • Use droppedspan outcome in service destination aggregation 10592
  • metrics: Count context.DeadlineExceeded as timeout 10594
  • metrics: Count GRPC codes.Canceled as timeout 10605
  • Remove global labels from aggregations for RUM services 10624
  • Remove redundant aggregation groups on faas.coldstart in transaction metrics 10635
APM version 8.7.0
edit

View commits

Breaking Changesedit
  • transaction.failure_count has been removed. transaction.success_count type has changed to aggregated_metric_double 9791
  • transaction.success_count has been moved to event.success_count 9819
  • Stop indexing transaction metrics to metrics-apm.internal 9846
  • Stop indexing span destination metrics to metrics-apm.internal 9926
  • apmserver.aggregation.txmetrics.overflowed metric has been renamed to apmserver.aggregation.txmetrics.overflowed.total 10330
  • Elasticsearch sourcemapping credentials require access to the .apm-source-map index 9722
  • Change APM Server default host to 127.0.0.1 9877
Bug fixesedit
  • We no longer unconditionally block shutdown for 5 seconds when tail-based sampling is enabled 9838
  • Fix a panic caused by accepting new grpc connections during shutdown 10013
  • Fix goroutine leak in libbeat final processor 10247
  • Fix servicemetrics aggregator name typo 10022
Intake API Changesedit
  • Add request.id to span.context.http 9673
  • Add agent.activation_method metadata field 9850
Addededit
  • Metrics data streams now use synthetic source 9756
  • Translate otel log events to error events 10066 10128
  • Use Elasticsearch fetcher for agent remote configuration where possible 9720 10088 10115
  • Fetch sourcemaps from Elasticsearch 9722 10367 10360 10106
  • Add support for RHEL 9 10317
Transaction metric improvementsedit
  • Add event.success_count to transaction metrics 9819 10165
  • Add transaction.duration.summary to transaction metrics 9973 10214
  • Introduce overflow buckets to limit cardinality and scale the max groups based on available memory 9856 10157 10362 9911 10012 10190
  • Add overflow metrics collection 10330
  • Improve APM UI query performance by producing 1m, 10m and 60m aggregation intervals for transaction metrics. Store them into hidden data streams metrics-apm.transaction.${interval} 9846 10070
  • Reduce memory footprint of transaction aggregations 10178
Service transaction metric improvementsedit
  • Introduce overflow buckets to limit cardinality and scale the max groups based on available memory 9856 10157 9911 10012
  • Add overflow metrics collection 10330
  • Rename service metrics to service transaction metrics 10095 10110
  • Improve APM UI query performance by producing 1m, 10m and 60m aggregation intervals for service metrics. Store them into hidden data streams metrics-apm.service.${interval} 9960 10070
  • Add service.language.name 10111
  • Add histogram support 9864 10214 10127 10216
  • Always enable metrics aggregaton and remove config option 10060 10109
  • Reduce memory footprint of service transaction aggregations 10178
Service destination metric improvementsedit
  • Introduce overflow buckets to limit cardinality 10069
  • Add overflow metrics collection 10330
  • Set _doc_count for service destination metrics 9931
  • Add global labels to service destination metrics 10056 10297
  • Improve APM UI query performance by producing 1m, 10m and 60m aggregation intervals for service destination metrics. Store them into data streams metrics-apm.service_destination.${interval} and make them hidden for non-default intervals 9926 10070
Introducing Service summary metric improvementsedit
  • Introduce service summary metrics 10111
  • Introduce overflow buckets to limit cardinality 10061 10302
  • Add overflow metrics collection 10330
  • Improve APM UI query performance by adding an always enabled service summary metrics aggregator which produces 1m, 10m and 60m aggregation intervals. Store them into data streams metrics-apm.service_summary.${interval} and make them hidden for non-default intervals 10029 10060 10070
  • Add service.language.name 10111