As you evolve your IT infrastructure with vSAN, staying on top of troubleshooting strategies will ensure you are prepared for those rare cases where an application or IT resource requires attention. By being able to identify issues quickly you can ensure your applications are functioning optimally and infrastructure upgrades are smooth. On May 30, Bradley Mott, will explore the following troubleshooting scenarios and more in the Getting More Out of vSAN: Troubleshooting webcast
Why can’t I power on my VM?
The first thing we need to do in this type of scenario is ensure that all the components that contribute to a healthy vSAN datastore are available. With this in mind, the vSAN Health Service in the vSphere Web Client is always a good place to start! It gives us answers to questions like:
- Are all the Hosts in the cluster?
- Are all the disks in the cluster?
- Are all the Hosts/disks healthy?
- Is there something weird we’re missing?
Answering these questions can help us understand why we’re seeing an outage. Most likely when a VM cannot power on we’ve either exceeded our Failures to Tolerate (FTT) policy in one way or another (such as hardware failure, network partition, logical segmentation, etc.), or we’ve run low on available space in the cluster. We’ll touch on these topics, and how to know!
Why won’t my host enter Maintenance Mode?
Similar to the first question we’re looking into, we should first confirm that the cluster is contributing all the resources we expect. From there, Maintenance Mode performs additional checks that need to be satisfied. By moving a host to Maintenance Mode, we’re taking away a percentage of the cluster’s resources that we need to consider. If all things are equal, and we have four hosts, putting one host in Maintenance Mode means we’re losing 25% of available resources. Do we have that amount of overhead available? There are, of course, some caveats surrounding objects (orphaned, inaccessible, and unhealthy), and we’ll discuss these in the live session.
How should I plan for host reboots, maintenance, upgrades, and changes?
Many of our customers are familiar with upgrading a vSphere environment, and how easy the process has become over the years! Products like vSphere Update Manager can roll through clusters in a snap, and make updates almost automatic. With vSAN, there are a few additional considerations, including if we want to potentially degrade data redundancy (Ensure Accessibility), provide redundancy during our maintenance (Full Migration), or if we prefer a full outage (with acceptable down time) during the upgrade to avoid resync/rebuild times in the middle.
Join us to learn about these scenarios as well as other topics including networking changes that may impact the cluster and VMware’s “best practices” approaches to the “speed versus redundancy” question during the upgrade process.
Make sure to register for the webcast and get your troubleshooting questions answered.