There are many application teams doing performance testing, and they need to collect the metrics from the Node.js agents.
After a recent restart of the APMIA agent, the Node.js agents are not connecting to the Node.js Collector in the Infrastructure Manager.
Everything was working fine and all the agents had been connecting before. No changes were made on any of the components (MOM, EM Collector, Node Collector, Node.js agents).
The Node.js agents are at version 1.10.83. The APMIA, MOM and EM Collector are at version 10.7.0.197.
The Node.js applications are running in OpenShift Containers.
Release : 10.7.0
Component : APM Agents
Customer figured out and resolved the issue.
Someone ran an Ansible Playbook on the server around 2pm on 6/9/20 that corrupted the /etc/resolv.conf and put some old nameservers in the file.
I found some entries in the /var/log/messages and /var/log/yum.log that showed some work was done around 2pm. I was able to pull a graph of the established connections from the IA that indicated that the agents started disconnecting at around 2pm.
The UNIX team corrected the file and the agents have reconnected.