Security monitoring tools (such as Microsoft Defender for Endpoint) flag the CA Release Automation (Nolio) Agent for initiating unexpected outbound traffic. Observed behaviors include:
NolioAgent process (e.g., /opt/nolio/jre/bin/NolioAgent).The outbound communication is not initiated by the Nolio Agent’s native application logic. Instead, it is caused by a third-party APM agent (Application Performance Management - e.g., Dynatrace OneAgent) that has been injected into the Nolio Agent's Java Virtual Machine (JVM).
When the monitoring agent is active within the Nolio process:
Example for Dyntrace Oneagent :
- Following Linux command
lsof | grep -i 'Nolio' | grep -i 'Dynatrace'
returns some Dynatrace OneAgent files indicating that it is injected in the NolioAgent java process
- In /var/lib/dynatrace/oneagent/agent/config/deployment.conf there is following line indicating the Auto Injection is enabled
AutoInjectionDisabled=false
- In /var/lib/dynatrace/oneagent/agent/config/ruxitagentproc.conf the agent Monitoring mode value is defined :
[General]
...
agentMonitoringMode FULL
or
agentMonitoringMode CLOUD_INFRASTRUCTURE_MONITORING
This behavior is expected in environments where the Nolio Agent is being monitored by an APM solution. To resolve the security flags, perform the following:
-agentpath or ruxitagent).agentMonitoringMode in ruxitagentproc.conf. Modes like FULL or CLOUD_INFRASTRUCTURE_MONITORING will result in different traffic patterns./var/log/secure or journalctl -u sssd) to confirm if Port 88 traffic is related to standard system Kerberos ticket requests.