Datadog/Datadog Agent

Releases15Avg5/moVersionsv7.74.1 → v7.78.0

Agent

Prelude

Release on: 2025-12-01

Please refer to the 7.72.4 tag on integrations-core for the list of changes on the Core Checks

Datadog Cluster Agent

Prelude

Released on: 2025-12-01 Pinned to datadog-agent v7.72.4: CHANGELOG.

Enhancement Notes

Update Cluster Agent to use latest schema for DatadogPodAutoscaler CRD.

Agent

Prelude

Release on: 2025-11-26

Please refer to the 7.72.3 tag on integrations-core for the list of changes on the Core Checks

Bug Fixes

Remove unused trace-loader binary from the heroku build to avoid package size increase

Datadog Cluster Agent

Prelude

Released on: 2025-11-26 Pinned to datadog-agent v7.72.3: CHANGELOG.

Bug Fixes

Fix a memory leak in the orchestrator check when the Cluster Agent is deployed with multiple replicas and leader-election enabled.

Agent

Prelude

Release on: 2025-11-12

Please refer to the 7.72.2 tag on integrations-core for the list of changes on the Core Checks

Bug Fixes

gpu: fix a bug where the gpu.sm_active and gpu.process.{sm_active,dram_active,encoder_utilization,decoder_utilization} metrics were emitting values lower than expected.
Fix regression in 7.72.0 where the install script on Windows (Install-Datadog.ps1 and datadog-installer-x86_64.exe) does not reinstall the Agent if the same version was just uninstalled.

Datadog Cluster Agent

Prelude

Released on: 2025-11-12 Pinned to datadog-agent v7.72.2: CHANGELOG.

Agent

Prelude

Release on: 2025-11-06

Please refer to the 7.72.1 tag on integrations-core for the list of changes on the Core Checks

Upgrade Notes

Upgraded the WDK version used by ddprocmon.

Bug Fixes

Fixed a regression with Prometheus scraping for Istio-injected containers introduced in version 7.70.
Fixed potential page fault and BSOD when trying to read environment variables from Windows user mode processes.
In Agent v7.69.0, the Autodiscovery filtering mechanism began applying filters defined in container_include to metrics and logs collection. This reverts the behavior to respect only the product specific filters.
Fixes regression affecting compatibility for versions of macOS older than 13.3 (involving the libz library).

Datadog Cluster Agent

Prelude

Released on: 2025-11-06 Pinned to datadog-agent v7.72.1: CHANGELOG.

Agent

Known Issues

This version does not work properly on versions of macOS older than 13.3, due to a dependency (libz) issue. There is currently no workaround and the recommendation at this time is to downgrade to Agent v7.71.2 or upgrade to v7.72.1 when it becomes available.

Prelude

Release on: 2025-11-03

Please refer to the 7.72.0 tag on integrations-core for the list of changes on the Core Checks

Upgrade Notes

The Agent's embedded Python has been upgraded from 3.12.11 to 3.13.7
The PowerShell install script Install-Datadog.ps1 now executes a different datadog-installer-x86_64.exe subcommand.

If you always download a fresh Install-Datadog.ps1, then no changes are needed.

If you have a cached or modified copy of Install-Datadog.ps1, then you must merge the latest changes, or migrate to use the executable setup instead. See the features section for more information.

Install-Datadog.ps1 is now a light wrapper for downloading and executing datadog-installer-x86_64.exe. The rest of the setup responsibility has been moved into the executable.

New Features

The Agent now emits the kube_distribution tag based on labels in managed Kubernetes installations with values: eks, gke and aks.
Add the new metric kubernetes.pod.resize.pending to kubelet check to track pods that have pending resource resize request that can't be satisfied right now. Metrics has reason tag.
Print service details in the Agent configcheck verbose command output.
Add Windows DDOT OCI package to support Datadog OpenTelemetry
The trace-agent running in the Azure App Services Extension now adds function tags to the tracer payloads that it sends to Datadog. This allows the traces to be queried by the trace tags, which is particularly useful for standard filtering in the Datadog UI.
Add support of CIS Red Hat Enterprise Linux 10 Benchmark in CSPM.
Add support of CIS Ubuntu 24.04 Benchmark in CSPM.
Added Datadog OTEL agent as Windows service

This is part of the effort to add Datadog OpenTelemetry agent as a Windows service.
Adds support for UDP, ICMP, and TCP Network Path for Windows Client OSes using ddnpm driver.
Added new Windows Server 2025 container images:

Release tags:
- agent:7-ltsc2025 and agent:latest-ltsc2025
- agent:7-servercore-ltsc2025 and agent:latest-servercore-ltsc2025
- agent:7-ltsc2025-jmx and agent:latest-ltsc2025-jmx
- agent:7-servercore-ltsc2025-jmx and agent:latest-servercore-ltsc2025-jmx
Release candidate tags:
- agent:7-rc-ltsc2025
- agent:7-rc-servercore-ltsc2025
- agent:7-rc-ltsc2025-jmx
- agent:7-rc-servercore-ltsc2025-jmx
Version-specific release candidate tags (such as agent:7.X.Y-rc.Z-ltsc2025, agent:7.X.Y-rc.Z-servercore-ltsc2025) are also supported.

Updated the base multi-arch images to include Windows 2025 images alongside existing Windows 2019 and Windows 2022 variants.
The Agent can now be configured to restrict which secrets from Kubernetes can be by integration (only apply to secret using the k8s_secret@ prefix).

Three new settings have been introduced, allowing different levels of control (see datadog.yaml.example for more information).
- `secret_scope_integration_to_their_k8s_namespace`: limit containers to their own namespace.
- `secret_allowed_k8s_namespace`: limit containers to a set of predefined namespaces.
- `secret_image_to_handle`: explicitly list which secret can be accessed by which container image.
Added integration test to ensure correct population of service and source tags on the truncate log metric.
USM HTTP2 monitoring configuration now uses a tree structure for improved organization and consistency with other protocol configurations.

Configuration changes:
- service_monitoring_config.enable_http2_monitoring → service_monitoring_config.http2.enabled
- service_monitoring_config.http2_dynamic_table_map_cleaner_interval_seconds → service_monitoring_config.http2.dynamic_table_map_cleaner_interval_seconds
The previous configuration paths are deprecated but still supported for backward compatibility. Users will receive deprecation warnings and should migrate to the new tree structure. If both old and new configurations are present, the new tree structure takes precedence.

Environment variable changes:
- DD_SERVICE_MONITORING_CONFIG_ENABLE_HTTP2_MONITORING → DD_SERVICE_MONITORING_CONFIG_HTTP2_ENABLED
- DD_SERVICE_MONITORING_CONFIG_HTTP2_DYNAMIC_TABLE_MAP_CLEANER_INTERVAL_SECONDS → DD_SERVICE_MONITORING_CONFIG_HTTP2_DYNAMIC_TABLE_MAP_CLEANER_INTERVAL_SECONDS
The datadog-installer-x86_64.exe executable can now be used in place of the Install-Datadog.ps1 PowerShell script. The executable is versioned and will only install that version. For example, to install Agent 7.72.0, download and run https://install.datadoghq.com/datadog-installer-7.72.0-1-x86_64.exe. For more information, refer to the [in-app install instructions](https://app.datadoghq.com/fleet/install-agent/latest?platform=windows).

To install the latest stable Agent version, download and run https://install.datadoghq.com/datadog-installer-x86_64.exe. This executable file is also versioned and will only install one version. It is updated with each release.

To check the version, run .\datadog-installer-x86_64.exe version.

Enhancement Notes

Add resize policies for CPU and memory as tags for containers when they are explicitly set on workloads.
Fix config path patterns for Python 3.13 and add SpecifierSet error handling.
The cluster agent API now supports querying for node UIDs by name.
Add support for KSM custom resource definition metrics of type Info, enabling collection and forwarding of custom resource info metrics
Added tracking of check worker utilization. Utilization stats can be seen using the agent status command. In addition, the Datadog Agent will log a warning if a worker's utilization exceeds a threshold (95% by default).
Add support for upstream datadogextension to DDOT.
Adds kubernetes.statefulset.rollout_duration metric to ksm check.
Added a new 'process' type to the workload filtering system for future use cases.
Add --json and --pretty-json flags to the health command that will output the health status as JSON.
[APM] Add support for sql_obfuscation_mode=normalize_only. This mode configures the SQL obfuscator to only normalize the values, without obfuscating them.
Agents are now built with Go 1.24.9.
Windows: Cache event publisher metadata to avoid repeated expensive calls to EvtOpenPublisherMetadata.
The gcp.run.job.enhanced.task.ended enhanced shutdown metric for Cloud Run Jobs now includes the exit code as a tag.
Since version 7.67.0, the site is converted to a FQDN if it is one of Datadog's domains. A configuration option convert_dd_site_fqdn.enabled has been added to allow disabling this behavior.
Add support for oss datadog extension in converter.
Create TCP endpoints for multi-region failover.
Support for custom queries based scaling in Datadog Pod Autoscaler
The agent config command now displays only user-configured settings by default, excluding default values. Use agent config --all to display all settings including defaults (previous behavior).
Added exponential backoff to host metadata collection at startup to increase early host-payload frequency and speed up node-level tag availability.
Enhanced fingerprinting configuration with file-specific overrides and improved fallback behavior. logs_config.fingerprint_config can now be set on a per-source and global basis, as either [disabled, line_checksum, byte_checksum].

File-specific configurations take precedence over global settings, with automatic fallback to global config when file configs are missing or incomplete.
The Datadog Installer now keeps YAML comments when updating the Agent configuration during installation. While comment text is preserved, its format may not be. Changes to whitespace, indentation, or blank lines may occur.
Migrates the WorkloadMeta filtering system to use the new filtering component. This ensures a consistent experience across all resources.
Allows NetPath on Windows without local firewall changes.
Network Path will now run multiple traceroutes and end-to-end probes for each endpoint.
Set the 'provider-kind' tag at the beginning of the static tag list.
For better troubleshooting, secret backend version information is now included in the agent secret command, flare output, and other metadata when secret_backend_type is configured and the backend executable supports the --version flag.
Avoid ReadString allocation in serverless-init log processing for improved efficiency.
Use zstd compression on logs in serverless-init for better performance.
dbm: add SQL obfuscation ReplaceBindParameter option to support obfuscating SQL bind parameters.
Update database monitoring payloads to use the default compression kind.
Enhanced Windows BSOD reports to include bugcheck arguments. Updated internal filter that drops irrelevant BSOD reports to look for a Datadog driver in any frames of the callstack instead of only the first frame.
Enhance Windows BSOD reports to include the crash call stack and the Agent version found in the crash dump.

Deprecation Notes

The Agent PowerShell module is deprecated. It was previously used only in private previews. Use the new [Remote Agent Management](https://docs.datadoghq.com/agent/fleet_automation/remote_management/?tab=windows) installer instead.
The --python argument is no longer used in the integrations subcommand.
USM: HTTP configuration flat keys are now deprecated in favor of tree structure format. The following configuration keys are deprecated: service_monitoring_config.enable_http_monitoring, service_monitoring_config.max_http_stats_buffered, service_monitoring_config.max_tracked_http_connections, service_monitoring_config.http_notification_threshold, service_monitoring_config.http_max_request_fragment, service_monitoring_config.http_map_cleaner_interval_in_s, service_monitoring_config.http_idle_connection_ttl_in_s, and service_monitoring_config.http_replace_rules. Use the new tree structure under service_monitoring_config.http.* instead (e.g., service_monitoring_config.http.enabled, service_monitoring_config.http.max_stats_buffered). The deprecated keys remain fully backward compatible, but the new tree structure takes precedence when both are configured.

Security Notes

Bumped Bouncy Castle dependencies to mitigate against CVE-2025-8885 and CVE-2025-8916 in FIPS images.

Bug Fixes

Bug fixes to make sure CRI-O collector populates container WLM image ID and repo digest consistently with containerd and docker implementations
Fixed a bug which causes the ecs_metadata.json file to not be generating when generating an Agent flare on ECS Fargate.
Fixes an issue which would result in some container.* metrics showing up without any container-related tags associated with them.
Utilize single origin information source for DogStatsD tag enrichment to avoid tag duplication.
Removes unused map that was being mutated in a read lock.
Fixes a bug in the JSON aggregator where the byte offset was not correctly calculated. This ensures no logs are re-collected when auto multiline detection is enabled when the Agent is restarted.
Fix an issue with the kube_static_cpus tag that resulted in multiple timeseries for the kubernetes.memory.limits metric.
Ignore ExternalData for tag enrichment in edge cases where ExternalData is not consistent with LocalData.
The following Kubernetes State Core check metrics now aggregate only scheduled Pods to reflect actually used resources:
- kubernetes_state.container.<cpu|memory>_requested.total
- kubernetes_state.container.<cpu|memory>_limit.total
Fix a panic that could happen in the kubelet check when a pod was deleted while collecting metrics.
Fixed the "Load more" button in the Agent GUI log page, which was failing to load additional log lines after the initial load.
Trace Stats with serverless-init are computed in the Agent only. Backend trace stats computation is disabled.
Fixed the search (Cmd+F) functionality in the settings page of the GUI.
Start the Agent service when "Configure" is clicked from the systray, if it is not running
Fixed an issue where certain autodiscovery errors were not shown in the Autodiscovery section of the agent status command.
Fixes an issue in the Windows NPM driver, where existing connections were incorrectly marked as timeouts when closed. Fixes handle cleanup issues in the Windows NPM driver. Removes overlapping filter in the Windows NPM driver.
Warns when Logs Agent log configuration is null and handles gracefully, instead of panicking.

Other Notes

Added the following agent telemetry metrics for debugging gRPC communication within the agent:

Metric Name	Type	Description
<span class="title-ref">grpc.request_count</span>	Counter	Total number of gRPC requests processed by the agent
<span class="title-ref">grpc.error_count</span>	Counter	Total number of gRPC errors encountered by the agent
<span class="title-ref">grpc.request_duration_seconds</span>	Histogram	Distribution of gRPC request latencies for agent communication
<span class="title-ref">grpc.payload_size_bytes</span>	Histogram	Distribution of payload sizes for gRPC calls within the agent

Add new metric origins to the Agent for BentoML, Hugging Face TGI, and IBM Spectrum LSF.
CPU and wall clock time collection in Python profiling is temporarily disabled to maintain compatibility with Python 3.13.
Add new metrics to the Agent's origins for Datadog Operator.

Datadog Cluster Agent

Prelude

Released on: 2025-11-03 Pinned to datadog-agent v7.72.0: CHANGELOG.

Bug Fixes

Fixes an Admission Controller issue where the UDS socket host path was hardcoded, causing mutated pods to miss APM or DogStatsD sockets when custom hostSocketPath values were set via Helm or the Operator.
Cluster Agent now scrubs secrets when generating flares and cluster check reports.

Agent

Prelude

Release on: 2025-10-15

Please refer to the 7.71.2 tag on integrations-core for the list of changes on the Core Checks

Bug Fixes

Stopped sending the example.metric metric to the /sketches endpoint

Datadog Cluster Agent

Prelude

Released on: 2025-10-15 Pinned to datadog-agent v7.71.2: CHANGELOG.

Agent

Prelude

Release on: 2025-10-08

Please refer to the 7.71.1 tag on integrations-core for the list of changes on the Core Checks

Bug Fixes

Fixes configuration parsing for extended container metric collection on the container core check.
Fix Remote Agent management configuration changes on Windows.
Fix race condition that in rare cases caused remote updates on Windows to fail without an error.

The Go context used when creating a detached subprocess was closed on exit. This triggered an async action to kill the detached subprocess, which in rare cases completed before the process exited.

Datadog Cluster Agent

Prelude

Released on: 2025-10-07 Pinned to datadog-agent v7.71.1: CHANGELOG.

Agent

Prelude

Release on: 2025-10-01

Please refer to the 7.71.0 tag on integrations-core for the list of changes on the Core Checks

Upgrade Notes

pkg/opentelemetry-mapping-go/otlp/metrics: Add interval argument to Consumer methods. All intervals passed as of this change will be 0, which is equivalent to the previous behavior. [#725](https://github.com/DataDog/opentelemetry-mapping-go/pull/725)
Upgrade go-sqllexer to v0.1.8

New Features

Added eBPF support for the prctl syscall. The following fields are now collected:
- `prctl.option`: The option value used in the syscall.
- `prctl.new_name`: If the option is PR_SET_NAME, the new process name.
- `prctl.is_name_truncated`: If the option is PR_SET_NAME, indicates whether the retrieved name was truncated. A metric also tracks the number of truncated names.
The infraattributes processor included in DDOT pipelines can now determine a missing container.id attribute based on (in order of priority):
1. process.pid;
2. datadog.container.cgroup_inode (inode number of the container's cgroup);
3. k8s.pod.uid, k8s.container.name, and optionally datadog.container.is_init.
This PR adds the HTTP Method and HTTP Endpoint to the stats protobuf file for the resource renaming feature.

Enhancement Notes

DogStatsD emits a new metric, metric_type_count, that tracks total gauges, counters, etc.
Remote tagger is now enabled by default in Datadog cluster check runners, resulting in improved tag support and coverage for metrics from checks such as kube_state_core when configured to run in cluster check runners.
Improved logs destination warning log with additional endpoint metadata such as full url, track type, EvP category, and origin.
Displays the partially redacted API key to each endpoint on the Logs Agent status page.
Add debugger v2 intake proxy.
Add scrubber to JMX integration config and metadata.
Add the new metric kubernetes.pod.terminating.duration to kubelet check to track pods that are stuck in the termination phase. If deletionTimestamp is set to a time in the future,`kubernetes.pod.terminating.duration` is only emitted when the current time reaches the time set in deletionTimestamp.
Adds kubernetes.deployment.rollout_duration metric to ksm check.
On Linux, the Agent is now able to bind to ports below 1024 by default, simplifying common logs, SNMP, and integrations configurations. This is done by setting AmbientCapabilities=CAP_NET_BIND_SERVICE in the Agent systemd units.

If you want to disable this, create a systemd drop-in file (e.g., /etc/systemd/system/datadog-agent.service.d/disable-capability.conf) with: ` [Service] AmbientCapabilities=`.
Added kube_argo_rollout tag for workloads controlled by Argo Rollouts.
Agents are now built with Go 1.24.6.
APM: Add new configuration "DD_APM_CLIENT_STATS_FLUSH_INTERVAL" to adjust client stats flushing interval. This can be used to test if longer flushing intervals can improve stats processing throughput.
Consistently retrieve the socket protocol (sk_protocol) and include it in the key for the flow PID map. This change ensures that the flow PID map uses a consistent key format that includes the socket protocol, improving the accuracy and reliability of flow tracking in network operations.
Increase the Logs Agent default max message payload size from 256 KB to 900 KB.
Install-Datadog.ps1 now checks that the downloaded datadog-installer-x86_64.exe has a valid Datadog code signature.
gather and expose kubernetes state metrics for init containers. gather and expose init containers resources requests and limits when container is set with restartPolicy=Always.
Add a new tag kube_static_cpus on kubernetes CPU metrics that indicates whether a container is using dedicated cores.
Adds a disable field to interface_configs to disable metric collection for selected interfaces.
Add RUM support to the OpenTelemetry mapping library.
DDOT now supports sending host metadata from OTel resource attributes. This is required for Gateway support. See https://docs.datadoghq.com/opentelemetry/mapping/host_metadata/ on how to configure the attributes.
The non-DDOT OTLP ingestion pipelines for traces and metrics now include the infraattributes processor, enabling more complete Kubernetes tag enrichment.
pkg/opentelemetry-mapping-go: Sync with upstream library through [commit 7c614b196a40f883502e86e195e89c233b4c1006](https://github.com/DataDog/opentelemetry-mapping-go/commit/7c614b196a40f883502e86e195e89c233b4c1006).
pkg/util/quantile: Sync with upstream library through [commit 7c614b196a40f883502e86e195e89c233b4c1006](https://github.com/DataDog/opentelemetry-mapping-go/commit/7c614b196a40f883502e86e195e89c233b4c1006).
Adds complete workloadmeta process-entity information to workload-list command logs (datadog-agent workload-list)

Deprecation Notes

Remove the deprecated routing processor from DDOT. Use the routing connector instead.
Logging to syslog with TLS is no longer supported.

Bug Fixes

Fixed an issue during Agent upgrade where all datadog-prefixed packages were treated as integrations, preventing non-integration libraries from being reinstalled.
De-duplicates service checks when auto-discovery is enabled with the docker listener. Ensuring pod annotated services only have one check and ignores configuration from the default file source.
Restores the ability to update the trace-agent in the Azure App Services Extension by removing the remote tagger requirement. The remote tagger is disabled when running in AAS Extension.
Upgrade datadog-traceroute to v0.1.4.
Add a safety check before calculating the container.memory.working_set metric to prevent an inflated, incorrect result (16 EiB).
Remediate duplicated logs when using the native EKS Fargate logging method.
Fix profiling billing when using Azure App Services.
Fix potential deadlock when querying sub-processes configuration. When other processes are in a failed state, the Agent HTTPS timeout would not be taken into account, causing it to wait forever. This would block the creation of the status page and cause missing information in flares.
Install-Datadog.ps1 now displays error messages when run in environments without a proper console, such as PSRemoting or PowerShell ISE.
Fixes a bug in the kubelet check where the first seen eviction would not result in the kubernetes.kubelet.evictions metric being emitted.
Fix redis span obfuscation for HSET command, as it allows multiples fields to be set at once since redis 4.0.0
Store previously seen images in workloadmeta, and only query the image status endpoint for new images. Previously, if image collection for CRI-O was enabled, the Agent would query the image status endpoint for every image, on every check. This was an expensive network request and would log verbose output to journald.
Fixed could not create SSL context: library has no ciphers error in the postgres integration for the FIPS version of the Agent.
Remove the call to the API server resource type cache when the kubernetes_state_core check is running on the node Agent, as the cache is not initialized.
Adds filtering for metrics in the kubelet summary provider that are excluded by the Autodiscovery exclusion annotations.
Remote Agent updates on Windows now consider MSI exit codes 3010 and 1641 to indicate success
Remote Agent updates on Windows now use the /norestart MSI option to prevent Windows from automatically rebooting the host in rare cases when files are in use.
pkg/opentelemetry-mapping-go/otlp/metrics: Correctly treat summary counts as cumulative monotonic sums instead of cumulative non-monotonic sums. [#635](https://github.com/DataDog/opentelemetry-mapping-go/pull/635)
pkg/opentelemetry-mapping-go/inframetadata: Send host meta on new hosts. [#652](https://github.com/DataDog/opentelemetry-mapping-go/pull/652), [#655](https://github.com/DataDog/opentelemetry-mapping-go/pull/655)
pkg/opentelemetry-mapping-go/otlp/metrics: Fix automatic intial point dropping when converting cumulative monotonic sum metrics. [#654](https://github.com/DataDog/opentelemetry-mapping-go/pull/654)
pkg/opentelemetry-mapping-go/inframetadata: Add support for host aliasing using resource attribute datadog.host.aliases. [#661](https://github.com/DataDog/opentelemetry-mapping-go/pull/661)
pkg/opentelemetry-mapping-go/otlp/metrics: Improve accuracy of exponential histogram conversion when count is 1. [#696](https://github.com/DataDog/opentelemetry-mapping-go/pull/696)
pkg/util/quantile: Improve accuracy of exponential histogram conversion when count is 1. [#696](https://github.com/DataDog/opentelemetry-mapping-go/pull/696)
pkg/util/quantile: Omit empty buckets when converting exponential histograms. [#699](https://github.com/DataDog/opentelemetry-mapping-go/pull/699)
pkg/opentelemetry-mapping-go/otlp/metrics: Do not consume host or tags for ResourceMetrics that only contain APM metrics. [#721](https://github.com/DataDog/opentelemetry-mapping-go/pull/721), [#722](https://github.com/DataDog/opentelemetry-mapping-go/pull/722), [#724](https://github.com/DataDog/opentelemetry-mapping-go/pull/724)
pkg/util/quantile: Add a binary generator to the quantile package to facilitate DDSketch to OTLPHistogram conversion. [#738](https://github.com/DataDog/opentelemetry-mapping-go/pull/738)
pkg/opentelemetry-mapping-go/otlp/logs: Log attribute values of type bool, int, and double will be represented as such. Previously, they were incorrectly converted to string. [#740](https://github.com/DataDog/opentelemetry-mapping-go/pull/740)
pkg/util/quantile: Return an error on non-monotonic bounds for histogram metrics. Previously, this would create a panic. [#744](https://github.com/DataDog/opentelemetry-mapping-go/pull/744)
Resolved possible segfault when running zstd-compressed journald log collection on systemd versions >= 256.
Fixed truncation telemetry so it records service and source values correctly, by moving metrics recording to the tailer layer.
Include bugcheck data if available as part of Windows crash reports.

Other Notes

Add checks.delay and checks.runs telemetry metrics for the agent checks. This enables break down and analysis by check_name and state where applicable.
Add a metric counter to the Agent telemetry for hostname.drift_detected. This helps track whether hostname resolution issues occur during the Agent's lifetime.
Add a new telemetry for inuse_bytes, frees, and allocations to monitor rtloader memory usage and behavior.
Scrubber: Added support for additional sensitive keys in configuration scrubbing:
- HTTP header-style API keys with "key" suffix: x-seel-api-key, x-goog-api-key, x-sonar-passcode
- HTTP header-style API keys with "token" suffix: x-consul-token, x-datadog-monitor-token, x-vault-token, x-vtex-api-apptoken, x-static-token
- HTTP header-style API keys with "secret" suffix: x-api-secret, x-ibm-client-secret, x-chalk-client-secret
- Exact key matches: cookie, private-token, kong-admin-token, accesstoken, session_token
These keys will now be automatically scrubbed from configuration files, logs, and other sensitive data to prevent accidental exposure of credentials.

Datadog Cluster Agent

Prelude

Released on: 2025-10-01 Pinned to datadog-agent v7.71.0: CHANGELOG.

New Features

Added support for gradual rollout in K8s SSI deployments. This feature is enabled if the following conditions are met:
- DD_ADMISSION_CONTROLLER_AUTO_INSTRUMENTATION_ENABLED=true and/or DD_APM_INSTRUMENTATION_ENABLED=true
- Remote config is enabled in the organization
- Using a Datadog registry (gcr.io/datadoghq, hub.docker.com/r/datadog, gallery.ecr.aws/datadog) This gradually rolls out access to new injector and tracer library releases to subsets of customers, with the ability to pause if issues are detected.

Enhancement Notes

Collect Datadog custom resources in the orchestrator check.
Add the ability to collect Argo Rollouts and Karpenter custom resources by default.

Agent

Prelude

Release on: 2025-09-15

Please refer to the 7.70.2 tag on integrations-core for the list of changes on the Core Checks

Enhancement Notes

Add new scrubbers for API keys and auth tokens in HTTP headers and YAML files. keys added:
- authority
- cms-svc-api-key
- sec-websocket-key
- x-api-key
- x-auth
- x-auth-token
- x-dreamfactory-api-key
- x-lz-api-key
- x-rundeck-auth-token
- x-stratum-auth
- x-sungard-idp-api-key
- auth-tenantid
- x-vtex-api-appkey
- x-octopus-apikey
- cainzapp-api-key
- x-pm-partner-key
- x-rapidapi-key
- lodauth
- statuskey
- x-functions-key

Datadog Cluster Agent

Prelude

Released on: 2025-09-15 Pinned to datadog-agent v7.70.2: CHANGELOG.

Agent

Prelude

Release on: 2025-09-10

Please refer to the 7.70.1 tag on integrations-core for the list of changes on the Core Checks

Bug Fixes

gpum: fix patching of SystemD device permissions

Datadog Cluster Agent

Prelude

Released on: 2025-09-10 Pinned to datadog-agent v7.70.1: CHANGELOG.

Agent

Prelude

Release on: 2025-09-03

Please refer to the 7.70.0 tag on integrations-core for the list of changes on the Core Checks

Upgrade Notes

GPUM: GPU workload metadata and autodiscovery are now enabled by default. If you want to disable this functionality, add "nvml" to the autoconfig_exclude_features configuration flag in your datadog.yaml file.
Upgraded the WDK version used by ddnpm.
Upgrade ddprocmon to version 1.3.0. This version contains minor cosmetic changes to the ABI.
None. This change is backward-compatible and does not modify behavior unless users explicitly set new environment variables.

New Features

Windows: Add certificate chain validation to Windows Certificate Store integration.
Added agent remote-config reset command to reset the remote configuration state which can be useful in troubleshooting issues related to remote configuration.
GPUM: Add new configuration flag gpu.enabled to agent config to control the GPU Core-Check component, disabled by default.
Adds Cisco IPsec VPN tunnel metrics and options data collection to SNMP. This can be enabled and disabled using the collect_vpn config.
Support Cloud Run Jobs in serverless-init.
The Windows Agent install script (Install-Datadog.ps1) now supports setting host tags and enabling logs collection at install time via the DD_TAGS and DD_LOGS_ENABLED environment variables. This allows users to configure tags and logs without post-install edits to datadog.yaml.

Enhancement Notes

Add cluster_arn, service_arn, and task_arn tags to the ECS tasks.
Add container_instance_arn as a host tag when running on ECS EC2.
Add a script to manage IIS .NET APM instrumentation on Windows
Add macOS build for AArch64/ARM64.
Added garbage collection of the tmp directory to the datadog-installer.
The Datadog autoscaler now can apply vertical scaling recommendations to sidecar init containers.
Agents are now built with Go 1.24.6.
Expose additional cgroup memory metrics: active_anon, inactive_anon, active_file, inactive_file, unevictable, shmem, file_mapped, file_dirty, file_writeback, page_tables, refault_anon, refault_file. Metrics are disbaled be default. To enable them please modify conf.d/container.d/conf.yaml and flip extended_memory_metrics to true.
Add enhanced metrics for task started, ended, and duration for Cloud Run Jobs.
gpu: Renamed gpu.core.usage and gpu.memory.usage to gpu.process.core/memory.usage
If a path is specified in the URL in logs_config.logs_dd_url, use it as a prefix.
Add the datadog.snmp.requests metric, which tracks the number of SNMP requests sent by the Agent to devices.
Added detection and replacement of tags: blocks in datadog.yaml to ensure only top-level, uncommented, and non-nested tags are modified during install.
Add thumbprints and serial numbers as tags for certificates and CRLs to the Windows certificate store integration.
Remote Agent management on Windows now automatically retries when the MSI returns error 1601 (ERROR_INSTALL_SERVICE_FAILURE).
Remote Agent Management on Windows now automatically retries when an MSI custom action returns error 1601 (ERROR_INSTALL_SERVICE_FAILURE) and causes Error 1719. The Windows Installer Service could not be accessed.

Deprecation Notes

Remove deprecated serializer options.
- enable_stream_payload_serialization
- enable_service_checks_stream_payload_serialization
- enable_events_stream_payload_serialization
- enable_sketch_stream_payload_serialization

Bug Fixes

Fix a panic in Docker file log parsing when received messages are null equivalent.
When the Autodiscovery scheduler gets blocked, it triggers a health check failure (reported in the agent health command).
When the prometheus_scrape.enabled option is set to true and the pod has a prometheus.io/port annotation, containers must expose that port in their spec in order for the Agent to schedule openmetrics checks.
Fixes an issue where pending ECS tasks would temporarily remove the global tags ecs_cluster_name and cluster_arn.
Fix bug causing integrations configured via docker labels or K8s annotations to be overwritten in the agent so logs from integrations now work when configured using via docker labels or K8s annotations.
Fix noisy error logs when scrubbing large pod container commands.
Remote updates on Windows now quotes and escapes MSI property values. This fixes an issue that caused remote updates to fail when the Agent username contains whitespace, for example DDAGENTUSER_NAME=NT AUTHORITY\SYSTEM.
Remote Agent updates on Windows now handle computer name updates done after the initial install of the Agent. This allows the use of golden images to deploy remotely updatable hosts.
gpu: Java programs using deeplearning4j are now properly detected and monitored
gpu: fix container-to-device assignment for Docker containers
Data Streams Messages: Fix Unscheduling of kafka_consumer integrations that was leading to duplicate integrations running in parallel.
The Kafka messages feature is not working in some cases when using K8s labels because config is translated to yaml.
Load check tags onto the orchestrator check configured to be dispatched and run as a cluster check
APM: Fix issue where the probabilistic sampler would drop all OTLP traffic.
Kubernetes logs source timestamps are now validated. Invalid timestamps will not be passed.
The Agent no longer logs an error each time it fails to connect to system-probe through \\.\pipe\dd_system_probe. This log message was added in Agent 7.63.0. The error is still logged if it persists long enough, see [#35790](https://github.com/DataDog/datadog-agent/pull/35790). Each failed connection attempt still logs the error at debug level to aid troubleshooting.
Fixed an issue where DD_TAGS and DD_LOGS_ENABLED were silently ignored during install, requiring manual configuration edits.
Fixed an issue where metrics in the Windows certificate store integration are merged when multiple certificates have the same subject.
Fixed an issue where the Windows Event Log tailer could lose historical events (the 'amnesia bug') when it started with no prior bookmark and no events were processed immediately. The Agent now correctly seeds an initial bookmark from the most recent event upon startup, ensuring no events are missed on subsequent runs.

Other Notes

Add handling for dbm-health events in the event platform forwarder. These events are used by Database Monitoring integrations to report on Agent status.
Add a new metric counter to the Agent telemetry for transactions.errors. This will help to monitor the rate of empty or rejected sketch payloads sent to the backend.

Datadog Cluster Agent

Prelude

Released on: 2025-09-03 Pinned to datadog-agent v7.70.0: CHANGELOG.

Enhancement Notes

Created a new configuration option orchestrator_explorer.custom_resources.max_count that will allow customers to increase the number of custom resources they can collect in the Cluster Agent.
Time zones are now collected for CronJobs in the orchestrator check
Collect EndpointSlices by default.

Bug Fixes

Correctly set the VerticalPodAutoscaler mode to "Auto" when not provided
Fixed an issue that caused some Datadog Cluster Agent replicas to restart during the first install.
When using leases for leader election with the leader_election_default_resource option, the Datadog Cluster Agent no longer creates an unused config map.
Fix the cronjob.on_schedule_check service check of the kubernetes_state check for CronJobs that have a specific time zone.

Agent

Prelude

Release on: 2025-08-27

Please refer to the 7.69.4 tag on integrations-core for the list of changes on the Core Checks

Upgrade Notes

Upgrade sqllite to 3.50.4 on Windows.

Datadog Cluster Agent

Prelude

Released on: 2025-08-27 Pinned to datadog-agent v7.69.4: CHANGELOG.

Agent

Prelude

Release on: 2025-08-26

Please refer to the 7.69.3 tag on integrations-core for the list of changes on the Core Checks

Enhancement Notes

Agents are now built with Go 1.24.6.

Security Notes

JMXFetch upgrade to 0.50.0 to mitigate against CVE-2025-52999.

Bug Fixes

Fixes a bug in the Trace Agent that prevented DDOT from shutting down gracefully.
Fixed truncation telemetry so it records service and source values correctly, by moving metrics recording to the tailer layer.

Datadog Cluster Agent

Prelude

Released on: 2025-08-26 Pinned to datadog-agent v7.69.3: CHANGELOG.

Agent

Prelude

Release on: 2025-08-20

Please refer to the 7.69.2 tag on integrations-core for the list of changes on the Core Checks

Upgrade Notes

Upgrade libxml2 to 2.14.5.

Datadog Cluster Agent

Prelude

Released on: 2025-08-20 Pinned to datadog-agent v7.69.2: CHANGELOG.

Agent

Prelude

Release on: 2025-08-18

Please refer to the 7.69.1 tag on integrations-core for the list of changes on the Core Checks

Bug Fixes

Remote Agent updates on Windows now checks if the custom Agent password is available before updating the Agent. This prevents the update from failing later and leaving the host without an Agent installed.

This is only applicable to domain-joined hosts using a custom Agent username and password. See [Agent 7.66.0 upgrade notes](https://github.com/DataDog/datadog-agent/releases/tag/7.66.0) for more information.

Datadog Cluster Agent

Prelude

Released on: 2025-08-18 Pinned to datadog-agent v7.69.1: CHANGELOG.

Agent

Prelude

Release on: 2025-08-14

Please refer to the 7.69.0 tag on integrations-core for the list of changes on the Core Checks

Upgrade Notes

The cilium conntracker is now enabled by default in the system-probe, and now expects the /sys/fs/bpf to be mounted at /host/sys/fs/bpf in containerized environments. The conntracker, if enabled, will fail to load unless this mount is provided, with the log line "not loading cilium conntracker since cilium maps are not present" in system-probe's log file. Users who have enabled this feature can either upgrade to the latest helm chart or add this mount to their container

New Features

Adds additional information and data related to the setsockopt hook.
- Socket Information:
  - Socket type
  - Socket family
  - Socket protocol
- Filter Information
  - Disassembled filter
  - Filter hash
You can now set the JAVA_TOOL_OPTIONS that JMXFetch uses by setting the jmx_java_tool_options configuration option in the datadog.yaml config file. This allows you to pass additional JVM options to JMXFetch, such as memory settings or system properties.
Adding a TracerPayloadModifier to the Trace Agent.
pkg/trace/api: Container tags hash is returned as a response header of the info endpoint.
Added new config option include_ephemeral_containers to collect Kubernetes ephemeral containers. The option is disabled by default. When enabled, the Agent will report container.* and kubernetes.* metrics for ephemeral containers. It will also collect logs and schedule checks for ephemeral containers when configured to do so.
Data Streams Monitoring: Adds new feature allowing users to retrieve messages from Kafka topics.
Change collect_gpu_tags config flag to be enabled by default. Now the Agent collects an additional gpu_host host tag for all hosts that have Nvidia GPUs.
Added new processing rule to omit truncated logs from being sent to ingest
GPU: Add GPM collector for Hopper and newer NVIDIA GPUs
Adds VPN tunnels and route table data collection to SNMP. This can be enabled/disabled using the collect_vpn config.
The NTP check on Windows now discovers the primary domain controller (PDC) on domain-joined hosts when use_local_defined_servers is enabled. If the PDC is unavailable, it automatically falls back to registry-defined servers. Check now performs order-insensitive server list comparisons, reduces log noise, and avoids using itself as a time source when running on a domain controller.
[Preview] The agent can now connect to the AWS SSM, AWS Secrets, Hashicorp Vault and Azure Keyvault secret management solutions to resolve secrets without requiring a user provided binary. For this, two new settings are introduced: secret_backend_type and secret_backend_config. For more information see: https://docs.datadoghq.com/agent/configuration/secrets-management
Added support in DDOT for the datadogexporter.proxy_url configuration option. This allows users to specify proxy settings for DDOT with the collector configuration.
Windows: Add CRL monitoring to the Windows Certificate Store integration.

Enhancement Notes

The serverless-init build uses the new TracerPayloadModifier to add Function Tags to the _dd.tags.function tag of the Tracer Payload to support serverless trace tagging.
Agents are now built with Go 1.24.5.
The user is now able to specify which features they want enabled inside of the converter. Previously, the user would have to either enable or disable everything.
DDOT now uses zstd compression for logs by default.
If a check has both a Go and a Python version, the Go version now has priority by default. This change should not have any visible impact, but if needed, you can disable this configuration by setting prioritize_go_check_loader to false.
GPUM: the "status" command now returns status of the system-probe part of GPU monitoring
Added new DogStatsD configuration option "dogstatsd_flush_incomplete_buckets". When enabled, DogStatsD will flush all received metrics during shutdown, regardless of which time-interval based bucket they belong to.
Agent integration metadata payloads now include the JMX integrations.
Allow users to configure the HTTP timeout for the Logs Agent.
No longer have the Logs Agent fall back to TCP when configuring logs_config.logs_dd_url with a http(s):// prefix.
If the Oracle can_connect check is critical, also set the can_query check to critical.
Display the number of times each log processor has been used in the Logs Agent status endpoint.
Reduce binary size by removing the Sensitive Data Scanner (SDS) from the logs agent.
OTLP spans support db.namespace semantic and map to db.name for DBM support.
Generate a more detailed warning when the Logs Agent tailer limit is reached.
Improved the granularity of the Logs Agent pipeline monitor to track the capacity of each individual component of the pipeline.
Remote Agent management operations on Windows now attempt to force stop the Agent services if they do not respond to Service Control Manager requests.
Remote Agent management on Windows now automatically retries when the MSI returns error 1618 (ERROR_INSTALL_ALREADY_RUNNING).

Bug Fixes

Correctly respect the ecs_collect_resource_tags_ec2 variable when calling the ECS Agent. Start caching tags to reduce burden on the ECS Agent. Start logging error responses from the ECS agent.
Fix a panic in Docker streams log parsing when stream messages are truncated on transmission.
Fix the cgroup reader bug that would prevent the generic container check from sending metrics when the Agent encountered a permission error.
Fixes invalid logs compression error in DDOT, sets DDOT logs compression to gzip.
Add support for selecting the endpoint resolution method using advanced AD identifiers in Kubernetes endpoint check configurations defined in files or configmaps. This enables static pod check configurations to correctly resolve the endpoint by setting resolve method to "ip".
Fixed the serializer exporter for the OSS Collector, which was not setting the correct proxy variables when sending metric data.
Fixed Windows installer overwriting install_info from setup scripts. When using Fleet Automation setup scripts, the subsequent MSI installation now skips writing install_info via a new SKIP_INSTALL_INFO flag, preserving the original setup script installation method tracking.
Fix Jetson check to correctly parse the output of tegrastats for Orin boards.
Fix incorrect container.memory.kernel value when running with Kernel >= 5.19 and cgroupv2
Breaking change - Fixes the Oracle service name tag to be service_name instead of service. This corrects the conflict with the APM service tag. This is a breaking change for any users who had been relying on the service tag to be set to the Oracle service name. The service tag can still be set explicitly in the tags configuration if needed.
Metrics sent from the process check on the core agent now have the host tag.
GPU: fix a bug where the device assigned to a process could be wrong if it updates the CUDA_VISIBLE_DEVICES environment variable during runtime
GPUM: fix Kubernetes device allocation detection in Google Kubernetes Engine
The NTP check will no longer fail to start if the initial discovery of local NTP servers fails at agent startup.
Limit the HTTP timeout on startup to 5 seconds for the Logs Agent.
Prevent the process component from running in the cluster worker.
Removes an extra copy of agent.exe from the Windows container
Remote Agent management operations on Windows now attempt to restart the Agent services after failing to stop the services or uninstall the Agent.
Fix Cgroup namespace not properly detected in Workload Protection, leading to incorrect container ID resolution and misqualified detections.

Other Notes

Add a new metric to the Agent telemetry for the startup and running states. This will help us track the startup and running states of the Agent.
Transparent Huge Pages (THP) usage is now disabled by default in the System Probe and Security Agent. To re-enable their usage, set the system_probe_config.disable_thp or security_agent.disable_thp configuration options to false.

Datadog Cluster Agent

Prelude

Released on: 2025-08-14 Pinned to datadog-agent v7.69.0: CHANGELOG.

Enhancement Notes

The auto-instrumentation webhook supports labels and annotations as tags configuration. If any of the label or annotation mappings for the incoming pod correspond to Universal Service Tags (service, env, or version), the webhook will also add the corresponding UST environment variable to the pod (DD_SERVICE, DD_ENV, or DD_VERSION).
The autoinstrumentation webhook will now set a default security context for init containers if the pod is in a namespace with a restricted security context. This can still be overridden by setting the environment variable DD_ADMISSION_CONTROLLER_AUTO_INSTRUMENTATION_INIT_SECURITY_CONTEXT.
Collect agent version in orchestrator check.

Agent

Prelude

Release on: 2025-07-28

Please refer to the 7.68.3 tag on integrations-core for the list of changes on the Core Checks

Upgrade Notes

Upgraded JMXFetch to 0.49.9 <https://github.com/DataDog/jmxfetch/releases/0.49.9> which reverts a change that broke support for Java 7. See 0.49.9 <https://github.com/DataDog/jmxfetch/releases/tag/0.49.9> for more details.

Enhancement Notes

Agents are now built with Go 1.24.5.

Bug Fixes

The Agent MSI no longer fails when it is unable to read the version information from [MsiGetProductInfo](https://learn.microsoft.com/en-us/windows/win32/api/msi/nf-msi-msigetproductinfow).

When upgrading from an Agent version earlier than 7.56, a failure in MsiGetProductInfo, combined with an MSI rollback, may result in missing system drivers. If this occurs, repairing or reinstalling the Agent will restore the missing drivers. This issue is typically caused by a corrupted Windows Installer registry. To repair these issues, refer to the [Microsoft Program Install and Uninstall Troubleshooter](https://support.microsoft.com/en-us/topic/fix-problems-that-block-programs-from-being-installed-or-removed-cca7d1b6-65a9-3d98-426b-e9f927e1eb4d).

Datadog Cluster Agent

Prelude

Released on: 2025-07-28 Pinned to datadog-agent v7.68.3: CHANGELOG.

Agent

Prelude

Release on: 2025-07-21

Please refer to the 7.68.2 tag on integrations-core for the list of changes on the Core Checks

Bug Fixes

Fix an issue with the Agent pre-install script that caused integrations shipped with the Agent to be removed during an Agent upgrade.
Print the correct FIPS status for the Cluster Agent when running in FIPS mode.

Datadog Cluster Agent

Prelude

Released on: 2025-07-21 Pinned to datadog-agent v7.68.2: CHANGELOG.

Agent

Prelude

Release on: 2025-07-17

Please refer to the 7.68.1 tag on integrations-core for the list of changes on the Core Checks

Security Notes

Bump the secret-generic-connector side binary to 0.2.5

Datadog Cluster Agent

Prelude

Released on: 2025-07-17 Pinned to datadog-agent v7.68.1: CHANGELOG.

Agent

Prelude

Release on: 2025-07-10

Please refer to the 7.68.0 tag on integrations-core for the list of changes on the Core Checks

Upgrade Notes

Bump the Python version to 3.12.11
Change how attribute precedence is handled. All fields are now evaluated across both span and resource attributes, using the following order of precedence (from highest to lowest):
- datadog.* span attributes
- datadog.* resource attributes
- standard span attributes
- standard resource attributes

New Features

Add a port of the Windows integrations-core Python network check to Go. This version is disabled by default but can be enabled with use_networkv2_check in your configuration.
Add support for Autodiscovery for RDS Postgres and MySQL databases.
Windows: Add remote certificate collection for the Windows Certificate Store integration.
Add a System Probe module that will collect software inventory data from the host.
Added logs.truncated and associated aggregate tags into /comp/core/agenttelemetry/impl/config.go
Workload protection (CWS) can now generate events based on the setsockopt syscall
Added a new logs.truncated metric to the Agent that reports the number of logs truncated before being sent. This metric helps monitor log volume loss due to truncation and is tagged by service and source for better visibility.

Enhancement Notes

The agent configcheck --verbose command and flares now include a section that lists all collected configurations, both matched and unmatched. This addition aids debugging by revealing which configurations the Agent has detected.
Adds in newly supported ap2.datadoghq.com site to the MSI's GUI menu.
Individual integrations can now set their own auto multiline configurations, including adding custom samples for logs specific to that integration.
Allows RDS autodicovery to work with an empty tag list. If an empty tag list is provided, the autodiscovery will not filter instances based on tags, allowing all RDS instances to be discovered.
OpenTelemetry instrumentation scope attributes are now converted into log attributes.
Introduce a new sample configuration file, application_monitoring.yaml, to support the Hands Off config feature. This file is automatically placed under /etc/datadog-agent/ on Linux systems only. Users can manually edit the file to apply application monitoring configurations.
Agents are now built with Go 1.24.4.
ecs_cluster_name is added as a global tag when running on EC2.
Improve the memory efficiency of obfuscator key generation.
In OTLP metrics ingestion, the instrumentation_scope_metadata_as_tags option is now enabled by default. This means scope attributes are now added as tags to metrics. If you have too many unique values for instrumentation scope attributes, this may cause cardinality issues. To mitigate this, you can disable the behavior by setting datadog.metrics.instrumentation_scope_metadata_as_tags to false.
Orchestrator manifests will now be published with all tags present in their metadata counterparts.
Single Step Instrumentation now uses the Python tracer major version 3 by default.
Refactor the logs-agent auditor to utilize a more testable architecture.
Add Kind, ApiVersion, and NodeName to manifests. Add HostName to CollectorManifest.
Sensitive text from custom resources is now scrubbed from the manifest. If a field is sensitive, all values within that field are automatically redacted, ensuring that sensitive data is not exposed even in nested structures.
Update registry writer to not write atomically when Agent runs on ECS Fargate to reduce memory leak.
Updated Windows container image labels to align with Linux image labels for better OCI compliance. Added standard Open Container Initiative (OCI) labels including image source, revision, and version information.

Bug Fixes

APM: Fix an issue where the trace-agent could panic during shutdown trying to obfuscate a SQL payload.
APM: Fix an issue where trace-agent could panic with "send on closed channel" during shutdown.
Prevent Logs Agent registry entries from being removed prematurely when the log source is still active.
Fixed TCP retransmit counts by excluding TCP keep-alive packets. Also fixed potential IRQL corruption and memory corruption related to IPv6 filters.
APM: Reduce the log level of APM Traces Received log message to debug. These values are available via metrics so this log is mostly just noisy.
Factor dependent services into the timeout when stopping the Agent service on Windows. Operations such as the stop-service Agent subcommand and remote updates now wait longer for the Agent and its subservices to stop before reporting an error.
Fixed debug log message for detected locally defined servers in NTP check.
Fixes a panic in the checks collector that occasionally occurs when the Agent is shutting down.
Fixes Python integrations not being persisted after Agent uninstall. Enables persisting integration during fleet updates.
Fixes multiline stacktraces being split up into separate logs when serverless-init is installed in-process.
Windows Agent remote updates now submit the remote config task state to the backend. This reduces the time it takes for a remote update to complete.
Windows Agent installer now uses absolute path to msiexec.exe instead of PATH lookup, improving installation reliability
Fixes telemetry reporting in the Agent Install Script for Windows PowerShell on hosts using PowerShell version less than 6 and without Internet Explorer installed, such as on a Server Core installation.
The Datadog Installer service on Windows is now set to manual start. This prevents alerts from tools that monitor automatically started services, such as the Windows Server Manager Dashboard.
Fix a bug that resulted in some Orchestrator Kubernetes manifests losing the configured "extraTags".
Fix how the Live Process and Live Containers sets the hostname when running in an Agent that is running in AWS Fargate
Applies SQL obfuscation logic to OpenTelemetry db semantics. Specifically, db.statement and db.query.text values will be obfuscated along with resource name and sql.query, according to obfuscation settings in the Agent config: https://github.com/DataDog/datadog-agent/blob/1768f80e3f14d0d300b1276ae23ec7c8237dde4c/pkg/config/config_template.yaml#L1226-L1364
Ensure serverless deployments send logs with gzip compression.
Fix a rare panic that can occur when a log is unable to be written to a TCP-based unreliable endpoint.
Fixed a bug where the system.cpu.num_cores metric could be incorrect on certain Windows platforms.
Fixed Windows container image metadata to properly include build timestamps and version information.

Other Notes

Add Origins for DuckDB, Keda and Supabase
Add metric origins for the Windows Certificate Store integration.
Add metric origins for new integrations.
SystemD units are now written by .deb and .rpm package scripts during the installation process. They were previously part of the package archive. We do not expect this change to affect users.

Datadog Cluster Agent

Prelude

Released on: 2025-07-10 Pinned to datadog-agent v7.68.0: CHANGELOG.

New Features

The admission controller can now enable kubelet API logging in the injected agent sidecar.

Enhancement Notes

Added a new metric to expose the ksm kube_cronjob_status_last_successful_time metric. The name of the metric is kubernetes_state.cronjob.duration_since_last_successful.
Single Step Instrumentation now uses the Python tracer major version 3 by default.

Bug Fixes

Stop sending telemetry associated with a DatadogMetric when the object is deleted.
Fix a bug in the Kubernetes State Metrics (KSM) check where custom resource metrics were incorrectly named using the kubernetes_state.customresource.<name> pattern instead of the intended kubernetes_state_customresource.<prefix>_<name> format.
Fixes a bug in the admission controller webhook that caused volume mounts to be skipped when other webhooks injected init containers after our own volume mounts had been added.
Properly take into account the timeZone field of the CronJob objects in the kubernetes_state.cronjob.on_schedule_check service check.