-
Install APM Server in Program Files directory on Windows {pull}14905[14905]
-
Fix self instrumentation events inconsistency when dealing with request timeout {pull}15123[15123]
-
Drop and recreate tail-based sampling badger db after exceeding storage limit for configured TTL time {pull}15106[15106]
-
Add sampling.tail.discard_on_write_failure config (default=false) to opt in to data loss when tail-based sampling storage limit is reached {pull}15159[15159]
-
Surface config parsing error under EA managed mode by logging and marking EA input unit as failed {pull}14574[14574]
-
Remove unnecessary hot reload under EA managed mode when apm tracing config is nil {pull}14865[14865]
-
In this release we’ve introduced an image based on the hardened Wolfi image to provide additional security to our self-managed customers, and improve our supply chain security posture.
-
Track all bulk request response status codes {pull}13574[13574]
-
Fix a concurrent map write panic in monitoring middleware {pull}14335[14335]
-
Apply shutdown timeout to http server {pull}14339[14339]
-
Tail-based sampling: Fix rare gc thread failure after EA hot reload causing storage not reclaimed and stuck with "storage limit reached" {pull}14340[14340]
-
Support for Jaeger is now deprecated, and will be removed in a future release {pull}13809[13809]
-
APM Server will no longer retry an HTTP request that returned 502s, 503s, 504s. It will only retry 429s. {pull}13523[13523]
-
APM Server now supports emitting distributed tracing for its own operation when running under Elastic Agent, and adds support for configuring a sampling rate {pull}14231[14231]