Add a Device

From Zenoss Wiki
This is the approved revision of this page, as well as being the most recent.
Jump to: navigation, search

Manually Adding a Device in Zenoss is one of the first things a new Zenoss Admin will do.

When you add a device to Zenoss, Zenoss does two things:

Modeling (Discovery)
First, Zenoss will discover the device, which means it will connect to it and create a logical object model of the device as well as any system components that it finds. The device will be remodeled every 12 hours by default, which will allow Zenoss to auto-discover any new network services, network interfaces, local storage, installed software, etc. that should be monitored. In general, new components will be auto-monitored by default, saving you the hassle of manually updating Zenoss when your systems change over time.
Monitoring
In addition, once added, the device will be monitored continually, by default every 60 seconds via ICMP echo (ping), and every 5 minutes via SNMP/SSH. The monitoring process also involves the collection of performance data and can result in events being raised for the device to notify you of things that require your attention.

This combination of auto-discovery (modeling) and monitoring is the heart of the Zenoss monitoring paradigm.

Adding the Device

Go to the Infrastructure tab, and click on the Add a Device icon, which looks like a computer with a plus in it, and select Add a Single Device.

You'll be presented with a window allowing you to tell Zenoss about some aspects of the device Zenoss needs to know to model it. You will need to specify the following fields:

Name or IP Address
Specify the DNS name or IP address for the device (IPv6 is supported in Zenoss 4.) Note that if you use an IP address, it must be a static IP.
Device Class
The Device Class defines the types of discovery (modeling) as the default ongoing monitoring settings to use for the device. For Linux systems monitored with SNMP, choose /Server/Linux. If you will be using SSH to monitor a Linux system, choose /Server/SSH/Linux. For routers and switches, choose something appropriate from the /Network Device Class heirarchy.
Title
Use this field to specify a name for your device. This is particularly useful if you have entered a numeric IP address.
Model Device
If you have SNMP running on the remote system, or you have SSH authentication already configured in Zenoss, then keep this box checked. Zenoss will model ('discover') the device and any associated components like network interfaces immediately.

Bulbgraph.png Note: If you have not yet set up Zenoss with credentials for connecting to your remote device, see Prepare Remote Device for steps on how to do this.

Modeling the Device

Now return to your device in the tree. If you chose to model the device, then the device will appear in the Infrastructure Tab in a few moments, after Zenoss has connected to it and discovered it and associated components.

If you did not chose to model the device (possibly because you needed to prepare the remote device for monitoring), then you can manually tell Zenoss to model the device right now. This is also very handy if the hardware configuration of the device has changed and you want Zenoss to detect these changes right away. To (re)model a device, first select the Infrastructure Tab and select the device in the right-hand pane by clicking on its name. Then, select the Gear Icon in the lower left part of the screen and choose Model Device from the menu.

You will see the output from the manual modeling run in a pop-up window. Once the device modeler shuts down, close the window, and Zenoss will refresh, and the discovery process will be complete.

Performance Data

One of the first things people generally want to see after adding a device to Zenoss is a detailed graph of performance data collected from the device. But it's important to understand that while data collected during modeling (discovery) is displayed immediately in the user interface, Zenoss needs some time to collect data samples before useful graphs can be displayed. This is because performance metrics are collected during the ongoing monitoring that begins after the device is modeled.

Zenoss collects performance data via SNMP, SSH or your favorite Cloud API every 5 minutes by default as part of the ongoing monitoring process. This performance data is used to populate RRD performance graphs which are visible under the Graphs section of the Device detail pane, as well as for components that were identified (such as Interfaces, under Components in the device detail view) during the modeling run. It can take about 5-15 minutes for sufficient performance data to be collected to allow initial graph points to be displayed. Be patient and give Zenoss some time to monitor your device, and soon you'll have useful RRD performance graphs to view in the user interface :)

Going Further

Automating adding/removing devices using a curl script