Determining When You Want Events to Be Triggered


You will need to take the following steps to determine the events you want to have triggered:

  • Select the objects or variables.

  • Select the devices and interfaces.

  • Determine the trigger values for each object or object/interface type.

  • Determine the severity for the event.

Determining what triggers to use on which objects and for which devices and interfaces requires an assessment of the baseline data gathered. If you've been doing what we recommend in this book, you have done a fair amount of the work already. Chapter 2 helped you understand how to gather and analyze the baseline data. In Part II of this book, we provide recommendations for what objects to set trigger points on and what range of settings might make sense. You will need to tailor these recommendations for your particular network.

You need to establish trigger points or thresholds (possibly rising and falling) for each object of interest on every device of interest in your network and, for each interesting interface if the object has instances for each interface on the device. If you've taken our advice from Chapter 3, you grouped the devices in your knowledge base. You'll probably only need to determine triggers for each media type and each category of device you defined, instead of for each device and object on that device you are interested in.

An example is CPU utilization on your routers. You will want to set different thresholds for routers processing lots of packets through the CPU (process switching), such as your dial access routers versus your core routers that are able to switch most packets through hardware. If you group all your dial access routers (such as Cisco AS5x00) together, you will be able to select a single threshold that makes sense for all of them. Likewise, one threshold setting will do for all of your core routers (such as Cisco 7xxx).

As another example, you'll probably want to set different utilization thresholds for LAN and WAN networks. Most network administrators set fairly conservative thresholds for utilization on LANs because keeping the LAN performance crisp is usually important and cost-effective. However, WAN bandwidth is an expensive recurring charge and administrators will often choose to allow higher utilization percentages across these more expensive links. See the recommendations in Part II of this book for what thresholds make sense for each media type.

You can use a combination of the importance of the interface or category and the severity of the particular trigger to set the priority of the event and, therefore, what attention the event should receive.

The more you can group and categorize your devices and interfaces, the easier it will be to automate the configuration of events. Use your knowledge base as the resource for this information.

Now that you know what events you want to configure in your network, it is important to understand what events are already built into your network devices. You can then determine which ones you can use. The next section covers this in detail.



Performance and Fault Management
Performance and Fault Management: A Practical Guide to Effectively Managing Cisco Network Devices (Cisco Press Core Series)
ISBN: 1578701805
EAN: 2147483647
Year: 2005
Pages: 200

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net