Making the Move to OpenTelemetry Easier for AppDynamics Customers With Unified Agent Management

Modern applications depend on a growing number of agents to collect and export telemetry. As environments expand across hosts, services, and Kubernetes clusters, managing those agents manually becomes increasingly difficult.

Teams may face inconsistent versions, configuration drift, limited visibility, and uncertainty about whether changes have been applied everywhere they are needed. For AppDynamics customers, these challenges can make the move to OpenTelemetry feel like a large and disruptive migration project.

Unified Fleet Management is designed to simplify that transition by providing a consistent way to deploy, configure, monitor, and update observability agents across an organization.

A Gradual Path From AppDynamics to OpenTelemetry

Many AppDynamics customers operate thousands of language agents, Machine Agents, or Cluster Agents in production. Replacing these agents all at once would create unnecessary operational risk and require extensive coordination across application and infrastructure teams.

The migration strategy is based on updating and reconfiguring existing agents rather than requiring customers to uninstall and reinstall everything.

Customers can upgrade their existing AppServer agents to combined agents that support both AppDynamics and OpenTelemetry capabilities. Existing deployment tools, including AppDynamics Smart Agent, Ansible, Puppet, Chef, CI/CD pipelines, Helm, and Kubernetes Operators, can continue to play a role in the rollout.

Customers can then choose the operating mode that best matches their migration stage:

• AppDynamics mode for existing production workflows
• Splunk Observability mode for OpenTelemetry-based operations
• Dual mode for evaluation and comparison

Dual mode can help teams establish baselines and evaluate OpenTelemetry data before completing the transition. Once they are comfortable with the new telemetry path, customers can move toward standalone OpenTelemetry agents and manage them through Splunk Observability Cloud.

The OpenTelemetry Collector as a Foundation

The Splunk Distribution of the OpenTelemetry Collector is an important part of the migration path.

Collectors can run as agents or gateways and provide capabilities such as processing, buffering, retrying, and exporting telemetry. They also provide a consistent path for sending application and infrastructure data to Splunk Observability Cloud.

The management architecture uses the Open Agent Management Protocol, or OpAMP, to connect agents and Collectors to a central control plane. Telemetry continues to travel through the OpenTelemetry data path, while management communication uses a separate channel.

For configuration changes that require restarts, an OpAMP Supervisor can coordinate operations between the management service and the Collector on hosts.

Manage Agents With a Common Control Plane

Fleet Management is designed to support two core areas so far:

• Inventory management
• Configuration management

Inventory management provides visibility into agent health, status, and versions.

Configuration management exposes the effective configuration an agent is actually running with and enables configuration updates for individual agents or groups. This helps teams identify drift, validate changes, and standardize settings across environments.

Fleet Management avoids the need to perform manual verification checks or updates. It reduces operational overhead and users no longer need to perform full redeployments that are usually accompanied by a change management burden – hindering quick troubleshooting or updates.

Configure Observability in the Context of the Service

Central observability teams are often responsible for telemetry standards but may not own the application code or have access to every service environment.

A unified experience can allow these teams to configure capabilities such as:

• Call graphs
• Profiling
• Sampling rates
• Collector processors
• OpenTelemetry instrumentation settings

These changes can be made in the context of the service being monitored. The experience can also provide confirmation that the configuration was received and applied, including whether an agent restart is required. With Dynamic remote configuration for AlwaysOn Profiling, teams can adjust CPU and memory profiling settings without editing deployment configuration or restarting an application process <add link to profiling remote configuration blog>

This creates a clearer path from configuration intent to operational outcome:

1. Identify the service and associated agent.
2. Apply the desired configuration.
3. Restart the relevant component when required.
4. Confirm that the change was applied.
5. Monitor the resulting telemetry and agent health.

Support for the Way Teams Already Deploy

Organizations use different tools to manage hosts and applications.

Host-based environments may rely on Ansible, Puppet, Chef, Smart Agent, or CI/CD pipelines. Kubernetes environments may use Helm charts or Operators.

Over time, customers can combine existing deployment automation with centralized management through the UI and APIs. This allows teams to use automation for repeatable changes while using the management experience for investigation, validation, and targeted updates.

A More Manageable OpenTelemetry Transition

OpenTelemetry adoption is both a technology change and an operational change.

Teams need a way to evaluate new agents, compare telemetry, manage configuration, and make changes safely while services remain in production. They also need a migration path that accounts for existing AppDynamics investments and the realities of large, distributed environments.

Unified Fleet Management is designed to provide that path.

By combining existing deployment workflows, combined agents, the OpenTelemetry Collector, OpAMP, and Splunk Observability Cloud, organizations can move toward OpenTelemetry in stages while improving consistency and control across their observability environments.

Try Observability Cloud Today for Free

If you’re not an Observability Cloud customer already, get your free edition now. Try OpenTelemetry management features like Fleet Management and Config Builder, as well as the AI Agent Observability and the test the full power of Observability Cloud. No gated features. No credit card required. No timed trial.