Kubernetes 2.0 Container Agent

Enhancements

Automated Helm Uninstallation of Kubernetes 2.0 Agent

Previously, when you removed the Kubernetes 2.0 Agent from the OpsRamp portal, you needed to manually uninstall the Helm release from your Kubernetes cluster.

With this enhancement, the Helm release is automatically uninstalled as soon as the agent configuration is deleted from the portal—no manual Helm commands required.

See Install Kubernetes 2.0 Agent document for more information.

Log Collection with MaxTime Setting

You can now define a MaxTime parameter for log collection. This setting restricts how far back the OpsRamp Agent looks for modified pod logs.

For example, Setting MaxTime to 2 hours ensures only recently modified pod log files are processed, which helps reduce overhead and boost performance.

See Configure Logs document for more information.

Trace Service Name Update

The trace service name has been updated from opsramp-master-agent to opsramp-trace-service for improved service identification.

See Tracing Proxy document for more information.

Expanded Support for Prometheus and JMX Monitoring

Kubernetes 2.0 Agent now supports monitoring any workload compatible with Prometheus or JMX receivers.
To configure: Append /prometheus or /jmx to the workload name.

See Prometheus Receiver document for more information.

Added Support for OpenTelemetry JMX Receiver

We are integrated the OpenTelemetry Contrib JMX Receiver to collect JVM metrics such as:

  • Heap/Non-Heap Memory
  • Garbage Collection Stats
  • Thread Counts
  • Custom application metrics

This simplifies monitoring for Java-based services using standard OpenTelemetry protocols

See JMX Receiver document for more information.

Custom Path for etcd Certificates

You can now specify a custom file path for etcd certificates using the hostPath field in the metrics configuration.

If not specified, the Agent defaults to the standard certificate location based on your Kubernetes distribution.

Improved Kubernetes Event Monitoring

We have improved how the Kubernetes events are monitored and displayed:

  • Kubernetes events are now automatically linked to the corresponding resource in OpsRamp. If the resource doesn’t exist, the event links to the cluster.
  • The Host column in logs is now empty (to reduce confusion). A new ResourceName column shows the exact Kubernetes object (e.g., Pod).
  • Renamed the metric from KubeEvents to kubernetes_events for consistency.
  • The component format is now shown as objectUuid - eventReason, making it easier to trace the origin of events.

OpsRamp Agent

Enhancements

Environment Variables for Log File Paths

You can now use environment variables to define absolute paths for monitored log files (e.g., $filepath/agent.log).

Supported Platforms:

  • Linux: Reads variables from /etc/environment
  • Windows: Uses system-level environment variables

See Log Files document for more information.

Support for Yearly Scheduled Maintenance

OpsRamp Agent now allows you to schedule maintenance tasks on a yearly basis—ideal for automating long-term maintenance windows.

See Scheduled Maintenance document for more information.

Support for Long Passwords on Windows Non-Admin Agent Installation

You can now use longer and more secure passwords when installing OpsRamp Agent on Windows non-admin accounts to improve security.

Display of Windows Service Name in Resource Page

The V2 Resource Page now shows both the Service Display Name and the actual Windows Service Name. This makes it easier to identify and manage your Windows services by providing more detailed information in one place.

See New Resources document for more information.

Increased Output Size Limit for Process Automation

The output size limit for process automation scripts has been increased from 64 KB to 4 MB, allowing scripts to handle and return larger amounts of data.

Reporting Device Reboot Status After Patching

The OpsRamp Agent now reports whether a device requires a reboot after patching using the following new metrics:

  • lastRebootTimestamp - Timestamp of the device’s last reboot.
  • lastRebootInitiatedBy - Indicates if the reboot was triggered by patching or manually initiated.
  • lastRebootPatchIDs - Lists the specific Patch IDs if the reboot was caused by patching.

Bug Fixes

  • Fixed an issue with KVM monitoring when the KVM type is QEMU-KVM.
  • Resolved an issue of garbled characters appearing in script output for Japanese environments.