Enterprises with large and complex IT environments typically add and decommission numerous systems during an average week – anywhere from several dozen to hundreds. In doing so, from time to time, the unfortunate end result is an unplanned service outage. This is not that surprising since one of the most difficult activities to achieve without incident is removing equipment that other systems in the network rely on, such as a SAN (FC) switch.
When a SAN switch is shut down and disconnected, certain hosts in the network may lose connectivity to their underlying storage volumes and experience an outage. Other hosts may suffer from performance degradation / service disruption and newly generated single points of failure (that is, increased risk of downtime) due to deceased I/O paths. Such single points of failure may remain hidden until a fault in the network occurs and an outage results. Then you find yourself wondering how this could happen with all the redundancy that’s in place?!
What’s affected if you shut down a single switch in the SAN layer?
This can all be avoided with careful planning and impact analysis prior to the decommissioning process. One of AvailabilityGuard reports enables you to do just that. The What-If Impact Analysis report enables you to select a SAN (FC) switch (and other options such as storage systems) and review an analysis of the business processes that will be affected by decommissioning the switch. The report has a short business-oriented view summarizing the impact per business service followed by a more detailed section with information on how each and every object is impacted (host, storage volume, zone, etc.).
Here’s an example of the high-level business impact analysis: