One of the themes that influences the majority of our newsletters is that the wide spread movement to adopt virtualization and cloud computing has a profound affect on how we plan, design, implement and manage IT resources. In the next few newsletters, we will discuss one of the ways that these movements impact the management of the IT infrastructure.
One of the factors that complicate management in a traditional IT environment is that due to the way that IP was designed, there is not a single repository of routing information. This is an issue because routing tables are automatically updated and the path that traffic takes to go from point A to point B may change on a regular basis. These changes may be precipitated by a manual process such as adding a router to the network, the misconfiguration of a router or by an automated process such as automatically routing around a failure. In this latter case, the rate of change might be particularly difficult to diagnose if there is an intermittent problem causing a flurry of routing changes typically referred to as route flapping.
Issues such as route flapping can be classified as logical as compared to a device specific issue such as a link outage. We talked to an IT manager for network automation and performance at a Fortune 100 company on the topic of logical problems such as route flapping. He said that in his 40 years of experience that it is “very common” to have logical issues cause availability and performance problems. We also talked to a senior director of SaaS delivery for a major IT service provider. He agreed with the IT manager and stated that in his organization logical issues cause between 10% and 15% of availability and performance problems and that most of the logical issues are routing issues. He added that 90% of the time the logical issue doesn’t take them offline but that whether or not it takes them offline that “Finding the cause can be a challenging endeavor. It is not as easy as troubleshooting a failed WAN link.”
Routing issues are not much of a challenge for an organization that has a simple hub and spoke wide area network. They are a challenge for any organization that has a complex, meshed network. The next newsletter will describe how routing issues are becoming even more of a challenge.
