-
Docker container is not run as root by default. 21213
-
Read Fleet connection information from
fleet.
instead offleet.kibana.
. 24713 -
Beats build for 32Bit Windows or Linux system will refuse to run on a 64bit system. 25186
-
Remove the
--kibana-url
frominstall
andenroll
command. 25529 -
Default to port 80 and 443 for Kibana and Fleet Server connections. 25723
-
Remove deprecated/undocumented IncludeCreatorMetadata setting from kubernetes metadata config options 28006
-
The
/processes/<subprocess>
endpoint proxies to the subprocess’s monitoring endpoint, instead of querying its/stats
endpoint 28165 -
Remove username/password for fleet-server authentication. 29458
-
Fix rename *ConfigChange to *PolicyChange to align on changes in the UI. 20779
-
Thread safe sorted set 21290
-
Copy Action store on upgrade 21298
-
Include inputs in action store actions 21298
-
Fix issue where inputs without processors defined would panic 21628
-
Prevent reporting ecs version twice 21616
-
Partial extracted beat result in failure to spawn beat 21718
-
Use symlink path for reexecutions 21835
-
Use ML_SYSTEM to detect if agent is running as a service 21884
-
Use local temp instead of system one 21883
-
Rename monitoring index from
elastic.agent
toelastic_agent
21932 -
Fix issue with named pipes on Windows 7 21931
-
Fix missing elastic_agent event data 21994
-
Ensure shell wrapper path exists before writing wrapper on install 22144
-
Fix deb/rpm packaging for Elastic Agent 22153
-
Fix composable input processor promotion to fix duplicates 22344
-
Fix sysv init files for deb/rpm installation 22543
-
Fix shell wrapper for deb/rpm packaging 23038
-
Fixed parsing of npipe URI 22978
-
Select default agent policy if no enrollment token provided. 23973
-
Remove artifacts on transient download errors 23235
-
Support for linux/arm64 23479
-
Skip top level files when unziping archive during upgrade 23456
-
Do not take ownership of Endpoint log path 23444
-
Fixed fetching DBus service PID 23496
-
Fix issue of missing log messages from filebeat monitor 23514
-
Increase checkin grace period to 30 seconds 23568
-
Fix libbeat from reporting back degraded on config update 23537
-
Rewrite check if agent is running with admin rights on Windows 23970
-
Fix issues with dynamic inputs and conditions 23886
-
Fix bad substitution of API key. 24036
-
Fix docker enrollment issue related to Fleet Server change. 24155
-
Improve log on failure of Endpoint Security installation. 24429
-
Verify communication to Kibana before updating Fleet client. 24489
-
Fix nil pointer when null is generated as list item. 23734
-
Add support for filestream input. 24820
-
Add check for URL set when cert and cert key. 24904
-
Fix install command for Fleet Server bootstrap, remove need for --enrollment-token when using --fleet-server 24981
-
Respect host configuration for exposed processes endpoint 25114
-
Set --inscure in container when FLEET_SERVER_ENABLE and FLEET_INSECURE set 25137
-
Fixed: limit for retries to Kibana configurable 25063
-
Fix issue with status and inspect inside of container 25204
-
Remove FLEET_SERVER_POLICY_NAME env variable as it was not used 25149
-
Reduce log level for listener cleanup to debug elastic/beats#25274
-
Passing in policy id to container command works 25352
-
Reduce log level for listener cleanup to debug 25274
-
Delay the restart of application when a status report of failure is given 25339
-
Don’t log when upgrade capability doesn’t apply 25386
-
Fixed issue when unversioned home is set and invoked watcher failing with ENOENT 25371
-
Fixed Elastic Agent: expecting Dict and received *transpiler.Key for '0' 24453
-
Fix AckBatch to do nothing when no actions passed 25562
-
Add error log entry when listener creation fails 23482
-
Handle case where policy doesn’t contain Fleet connection information 25707
-
Fix fleet-server.yml spec to not overwrite existing keys 25741
-
Agent sends wrong log level to Endpoint 25583
-
Fix startup with failing configuration 26057
-
Change timestamp in elatic-agent-json.log to use UTC 25391
-
Fix add support for Logstash output. 24305
-
Do not log Elasticsearch configuration for monitoring output when running with debug. 26583
-
Fix issue where proxy enrollment options broke enrollment command. 26749
-
Remove symlink.prev from previously failed upgrade 26785
-
Fix apm-server supported outputs not being in sync with supported output types. 26885
-
Set permissions during installation 26665
-
Disable monitoring during fleet-server bootstrapping. 27222
-
Fix issue with atomic extract running in K8s 27396
-
Fix issue with install directory in state path in K8s 27396
-
Disable monitoring during fleet-server bootstrapping. 27222
-
Change output.elasticsearch.proxy_disabled flag to output.elasticsearch.proxy_disable so fleet uses it. 27670 27671
-
Add validation for certificate flags to ensure they are absolute paths. 27779
-
Migrate state on upgrade 27825
-
Add "_monitoring" suffix to monitoring instance names to remove ambiguity with the status command. 25449
-
Snapshot artifact lookup will use agent.download proxy settings. 27903 27904
-
Fix lazy acker to only add new actions to the batch. 27981
-
Allow HTTP metrics to run in bootstrap mode. Add ability to adjust timeouts for Fleet Server. 28260
-
Fix agent configuration overwritten by default fleet config. 29297
-
Allow agent containers to use basic auth to create a service token. 29651
-
Fix issue where a failing artifact verification does not remove the bad artifact. 30281
-
Reduce Elastic Agent shut down time by stopping processes concurrently 29650
-
Move
context cancelled
error from fleet gateway into debug level. 187 -
Update library containerd to 1.5.10. 186
-
Add fleet-server to output of elastic-agent inspect output command (and diagnostic bundle). 243
-
Update API calls that the agent makes to Kibana when running the container command. 253
-
diagnostics collect log names are fixed on Windows machines, command will ignore failures. AgentID is included in diagnostics(and diagnostics collect) output. 81 92 190 262
-
Collects stdout and stderr of applications run as a process and logs them. 88
-
Prepare packaging for endpoint and asc files 20186
-
Improved version CLI 20359
-
Enroll CLI now restarts running daemon 20359
-
Add restart CLI cmd 20359
-
Add new
synthetics/*
inputs to run Heartbeat 20387 -
Users of the Docker image can now pass
FLEET_ENROLL_INSECURE=1
to include the--insecure
flag with theelastic-agent enroll
command 20312 20713 -
Add
docker
composable dynamic provider. 20842 -
Add support for dynamic inputs with providers and
{{variable|"default"}}
substitution. 20839 -
Add support for EQL based condition on inputs 20994
-
Send
fleet.host.id
to Endpoint Security 21042 -
Add
install
anduninstall
subcommands 21206 -
Use new form of fleet API paths 21478
-
Add
kubernetes
composable dynamic provider. 21480 -
Send updating state 21461
-
Add
elastic.agent.id
andelastic.agent.version
to published events from filebeat and metricbeat 21543 -
Add
upgrade
subcommand to perform upgrade of installed Elastic Agent 21425 -
Update
fleet.yml
and Kibana hosts when a policy change updates the Kibana hosts 21599 -
Update
install
command to perform enroll before starting Elastic Agent 21772 -
Update
fleet.kibana.path
from a POLICY_CHANGE 21804 -
Removed
install-service.ps1
anduninstall-service.ps1
from Windows .zip packaging 21694 -
Add
priority
toAddOrUpdate
on dynamic composable input providers communication channel 22352 -
Ship
endpoint-security
logs to elasticsearch 22526 -
Log level reloadable from fleet 22690
-
Push log level downstream 22815
-
Add metrics collection for Agent 22793
-
Add support for Fleet Server 23736
-
Add support for enrollment with local bootstrap of Fleet Server 23865
-
Add TLS support for Fleet Server 24142
-
Add support for Fleet Server running under Elastic Agent 24220
-
Add CA support to Elastic Agent docker image 24486
-
Add k8s secrets provider for Agent 24789
-
Add STATE_PATH, CONFIG_PATH, LOGS_PATH to Elastic Agent docker image 24817
-
Add status subcommand 24856
-
Add leader_election provider for k8s 24267
-
Add --fleet-server-service-token and FLEET_SERVER_SERVICE_TOKEN options 25083
-
Keep http and logging config during enroll 25132
-
Log output of container to $LOGS_PATH/elastic-agent-start.log when LOGS_PATH set 25150
-
Use
filestream
input for internal log collection. 25660 -
Enable agent to send custom headers to kibana/ES 26275
-
Set
agent.id
to the Fleet Agent ID in events published from inputs backed by Beats. 21121 26394 26548 -
Add proxy support to artifact downloader and communication with fleet server. 25219
-
Add proxy support to enroll command. 26514
-
Enable configuring monitoring namespace 26439
-
Communicate with Fleet Server over HTTP2. 26474
-
Pass logging.metrics.enabled to beats to stop beats from adding metrics into their logs. 26758 26828
-
Support Node and Service autodiscovery in kubernetes dynamic provider. 26801
-
Increase Agent’s mem limits in k8s. 27153
-
Add new --enroll-delay option for install and enroll commands. 27118
-
Add link to troubleshooting guide on fatal exits. 26367 27236
-
Agent now adapts the beats queue size based on output settings. 26638 27429
-
Support ephemeral containers in Kubernetes dynamic provider. #27020 27707
-
Add complete k8s metadata through composable provider. 27691
-
Add diagnostics command to gather beat metadata. 28265
-
Add diagnostics collect command to gather beat metadata, config, policy, and logs and bundle it into an archive. 28461
-
Add
KIBANA_FLEET_SERVICE_TOKEN
to Elastic Agent container. 28096 -
Enable pprof endpoints for beats processes. Allow pprof endpoints for elastic-agent if enabled. 28983
-
Add
--pprof
flag toelastic-agent diagnostics
and anelastic-agent pprof
command to allow operators to gather pprof data from the agent and beats running under it. 28798 -
Allow pprof endpoints for elastic-agent or beats if enabled. 28983 29155
-
Add --fleet-server-es-ca-trusted-fingerprint flag to allow agent/fleet-server to work with elasticsearch clusters using self signed certs. 29128
-
Discover changes in Kubernetes nodes metadata as soon as they happen. 23139
-
Add results of inspect output command into archive produced by diagnostics collect. 29902
-
Add support for loading input configuration from external configuration files in standalone mode. You can load inputs from YAML configuration files under the folder
{path.config}/inputs.d
. 30087 -
Install command will skip install/uninstall steps when installation via package is detected on Linux distros. 30289
-
Update docker/distribution dependency library to fix a security issues concerning OCI Manifest Type Confusion Issue. 30462
-
Add action_input_type for the .fleet-actions-results 30562
-
Add support for enabling the metrics buffer endpoint in the elastic-agent and beats it runs. diagnostics collect command will gather metrics-buffer data if enabled. 30471
-
Update ack response schema and processing, add retrier for acks 200
-
Enhance error messages and logs for process start 225
-
Changed the default policy selection logic. When the agent has no policy id or name defined, it will fall back to defaults (defined by $FLEET_SERVER_POLICY_ID and $FLEET_DEFAULT_TOKEN_POLICY_NAME environment variables respectively). 29774 226
-
Add Elastic APM instrumentation 180
-
Agent can be built for
darwin/arm64
. When it’s built for bothdarwin/arm64
anddarwin/adm64
a universal binary is also built and packaged. 203 -
Add support for Cloudbeat. 179
-
Fix download verification in snapshot builds. 252
-
Add support for kubernetes cronjobs 279
-
Increase the download artifact timeout to 10mins and add log download statistics. 308
-
Save the agent configuration and the state encrypted on the disk. 535 398
-
Bump node.js version for heartbeat/synthetics to 16.15.0
-
Support scheduled actions and cancellation of pending actions. 393 419