Introduction
In environments where clients host Alaris Labs software on their own infrastructure, maintaining a constant connection with our monitoring system is crucial. Our custom Zabbix agent, known as tm_agent, plays a vital role in this process. This article provides guidance on what steps to take if you receive a notification about tm_agent losing connection with our monitoring server.
Understanding the Alert
When our monitoring system detects that tm_agent on a client's node (<hostname>) is unreachable for 10 minutes, it triggers an alarm in Zabbix:
tm_agentd on
<hostname>is unreachable for 10 minutes
Subsequently, clients are notified via email on the Contact Emergency list.
Immediate Actions for Clients
Upon receiving this alert, clients should perform the following checks:
SSH Access: Verify that the node is accessible via SSH. This is a preliminary check to ensure that the node itself is operational.
Port Check: Ensure that port 22031, which tm_agent listens on, is open. This port is essential for the agent's communication with the monitoring server.
Agent Data Transmission: Check that tm_agent is actively sending data to the monitoring server. This can be confirmed by reviewing the agent's operational logs.
- Routing check: Check the routing of packets on your infrastructure, make sure that there is network connectivity between the nodes and our monitoring servers.
Configuration File Checks The agent’s configuration file specifies the monitoring server details. Check the following files based on your infrastructure setup:
- For standalone setups:
/etc/alarislabs/tm/tm_agentd.conf(Server section)
If you have a cluster infrastructure installed, in particular, Pacemaker + Corosync, then you also need to check the cluster agent (tm_agentd_cluster), which is located only on the master node of the SMS Cluster:
- For cluster infrastructure:
/u02/configs/tm/tm_agentd.conf(applicable only on the master node of the cluster)
Cluster Infrastructure Specifics
In a cluster setup, two tm_agents operate on the master node: the cluster agent and the daemon agent. This dual-agent setup ensures both nodes in the cluster (master and slave) are monitored efficiently. It's crucial to verify that both agents (cluster and non-cluster) on the master node are functioning correctly.
NOTE: 2 agents will be running on the Master node (tm_agentd and tm_agentd_cluster). This is a normal situation.
Conclusion
Timely response to connectivity issues with tm_agent is crucial for maintaining effective monitoring of your hosted Alaris Labs software. By following these steps, clients can quickly diagnose and address common problems related to tm_agent connectivity.
Further Assistance
We can check traceroute, mtr, ping etc. to your servers upon request. Please contact Alaris Labs support for assistance.
It is important that if the operating system is administered by the client, then it is necessary to contact your system administrators or network engineers to resolve the issue.
AKBSMS - Alaris Knowledge Base
Related Questions:
What should I do if tm_agentd on my node is unreachable in Alaris Labs monitoring system?
How can I check if port 22031 is open for tm_agent communication?
What are the steps to verify SSH access to a node in Alaris Labs monitoring system?
Where can I find the tm_agentd configuration file in a standard setup?
What is the process for checking tm_agent on a master node in a cluster infrastructure?
How do I troubleshoot tm_agent connectivity issues in Alaris Labs Zabbix monitoring system?
What are common reasons for tm_agent losing connection with the monitoring server?
Where is the tm_agentd configuration file located in a cluster infrastructure?
What immediate actions should be taken if tm_agentd becomes unreachable?
How to identify and resolve tm_agent communication issues in Alaris Labs monitoring setup?
What role does port 22031 play in tm_agent's operation in Alaris Labs monitoring system?
What support options are available for persistent tm_agent connectivity problems in Alaris Labs system?
Link to this article: https://helpdesk.alarislabs.com/en/knowledge_base/article/308/category/144/