Agent settings define how enrolled agents connect to the platform, which serving endpoints they use, and how logging is configured. Settings are stored on the server and delivered to the agent at enrollment and at each subsequent sync. The agent always initiates communication — it contacts the serving endpoint on a regular sync cycle (default: every 10 minutes) to pick up pending messages from the server and to upload any results it has collected.
Managing agent settings
Agent settings are managed from Configuration > Infrastructure > Agent settings. To create or edit an agent settings entry, click Add or select an existing entry and click Edit. Updated settings take effect on enrolled agents at the next sync. Refer to Agent enrollment configuration for details on creating and managing agent settings entries.
Serving endpoints
The agent's serving endpoint is the address it contacts to communicate with the platform. Depending on your network topology, the serving endpoint can be either the backend server directly or a runner that proxies traffic to the server. In both cases the agent always initiates the connection — the server queues messages (such as inventory job assignments or updated settings) that the agent retrieves at the next sync.
When a runner is used as the serving endpoint, the runner's public-facing DNS name or IP address and port must be configured in agent settings. The runner's address is visible in the Details view of the runner entry under Configuration > Infrastructure > Runners.
Sync interval
The agent contacts the serving endpoint on a regular cycle to pick up pending messages from the server and to upload results. The default sync interval is 10 minutes. Actions that the server sends to the agent — such as a triggered inventory job — are therefore picked up and started within the next sync cycle. The sync interval can be adjusted in the agent settings entry.
TLS and certificate trust
When the runner is configured with an HTTPS endpoint, agents must trust the runner's certificate chain. This mirrors the requirement for other components that connect to an HTTPS runner endpoint. Ensure that the root certificate authority (CA) used to sign the runner's certificate is trusted on all devices running the agent. Refer to Appendix A: Maintaining the Public Key Infrastructure (PKI) for guidance on certificate management.
Multiple serving endpoints and failover
An agent settings entry can include multiple serving endpoints. The agent tries each endpoint in order and connects to the first one that responds successfully. If the primary endpoint becomes unavailable, the agent automatically switches to the next available endpoint without manual intervention. Configure multiple endpoints to improve resilience in environments where a single runner may be temporarily unreachable.
Logging levels
The agent and its scheduler component each write log files to the logs/ subfolder of the agent's application data folder. The log level controls how much information is written.
Available log levels, from least to most verbose, are: Error, Warning, Information, Debug.
|
Warning: |
Choose the log level that gives you the information you need without generating unnecessary log volume. For normal operations, Information is the recommended log level.