The goal here is to monitor a server running Slackware 13.
ITM agentless monitoring for Linux works using SNMP. The target systems needs to have the snmpd daemon running so ITM can get performance info from the system.
In this example Slackware has been configured to run snmpd in the default 161 udp port. Before running snmpd edit the following file:
The configuration file has several options and I am not going to discuss all of them here. If you're interested there's a lot of info in the following place: SNMPD manual
For now there's only one options that needs to be set: rocommunity
Set this option to whatever community value you want to use in order to get information from this server. You will need this value when configuring ITM agentless monitor. In this example I have used:
After configuring the snmpd daemon can be started. In Slackware the command is:
and in order to have it started within the server's start process you can run:
chmod +x /etc/rc.d/rc.snmpd
The next step is to configure ITM to monitor this server through SNMP.
In Tivoli Enterprise Portal (TEP) find a suitable Linux server to host the agentless monitor.
In this case I am using my own ITM server to host the monitor which will remotely contact the target server. This machine is called itm622.
Right Click in the server and select Add Managed System...
In the next screen select Agentless Monitoring for Linux...
Configure the agent with an instance name, port number and snmp version. Usually the default values (except for the Instace name) will be enough. Change them if required and click on Next
Now you need to input the community name you used in the snmpd.conf configuration file in the target system. In this case it will be tivoli. Confirm it and click on Next
On the next screen click in New to add a new monitored system. Fill in the name on which the server will be displayed on TEP and its IP Address or hostname so ITM can connect to it.
Click on Finish and the agent installation and configuration will start. After a couple of minutes TEP will show a message saying it needs to be updated:
Click on the button to see the new monitored system on the tree.
Click on the monitored system's name in order to see its information:
Using agentless technology the following information may be obtained:
- Overall processor, disk and memory utilization
- Logical disk utilization
- Virtual and physical memory utilization
- Network utilization and network cards
- Processes details
- Processor utilization real and over time
- System Summary
Agentless technology is useful not only for Operating Systems that do not support the ITM agent but it can also be used in the following scenarios:
- Basic OS info and key metrics may be enough for several monitoring needs
- Lightweight monitoring for small servers
- Standard protocols used in case of network restrictions
- Easy administration, roll-out and maintenance