One of the most powerful tools created between the Dell EMC and VMware joint engineering of the VxRail solution is VxRail Manager. The first thing to note is VxRail Manager wasn’t created to present a new management pane for daily tasks. We’ve realized that our customers know how to use vCenter and we don’t want to replace that skill set – that’d be silly, in my opinion. When we talk to customers about VxRail (and HCI in general), it’s about automation, simplicity and predictability. (Side note – VxRail 3.0 did have a workflow to create new VMs – it was removed from VxRail 3.5, for the reasons above – customers didn’t really want it.)
So, what does VxRail Manager do? It provides information about the cluster (mostly the same as what you see in vCenter) and it provides an interface for life cycle management of the hardware and software of the cluster, helps with support and provides physical and logical views of the cluster. All in all, it’s a tool to augment vCenter for some information as well as providing the automation for upgrades, adding nodes, adding drives and replacing drives.
Today, you log into VxRail Manager by accessing the IP address of the VxRail Manager VM. We use the vCenter SSO, so you don’t have to define/create new users for VxRail Manager access. When you first log in, you’ll see the dashboard:
The Dashboard provides a quick glimpse into the system. You’ll see the system health, a few of the recent community posts, support information and event history. This screen will also show if there are new nodes ready to be added to the cluster. Please excuse my “Error” status for system health – this is a lab box that has taken a beating and I didn’t reset some of the HW alerts from failure testing that we did.
The easiest way to navigate VxRail Manager is to go along the icons on the left hand side. As you can see, we have 5 different views we can look at. After the “Dashboard”, the next view is “Support”.
This is probably one of the benefits of VxRail Manager – from here, you can get a variety of support information, but there’s more that can be done. You can also open a support case via “Chat with Support” where you’ll have a IM support conversation with our support engineers, or you can open a service request and have someone reach out to you. Both are great/easy ways to open a support case. You also see links to get the VxRail Software downloads and you can natively search to VxRail Knowledge Base, if you’d like. If we had a support account associated with this cluster, you’d also see a few KB articles in the area below the KB search area.
The next tab is the events view:
In this view, you’ll see all the VxRail events that the system has logged. You can sort the events by ID, Severity, or Time. You can also search the events, if you’re looking for a specific one. When you select an event, you’ll see more details about that event and a KB link, if it’s applicable. One other thing about this event view is that starting with 4.5.200 and later code, we also inject the alerts into vCenter if they were events/alerts created by VxRail Manager. This helps to make sure that administrators have the alerts regardless of their preferred view.
The next tab is the “Health” tab. With this tab, you can gain insight to the system from two different approaches “logical” and “physical”. The default view is “physical”:
In the physical view, you will see in this G Series (2U, 4 Nodes) all 4 nodes (back view) and all 24 drives associated with the node. You can select one of the nodes to get all the info on that node. This is also where you’d go to add drive (cache and/or capacity) to a node. On the 1U, 1 Node (E Series) and the 2U, 1Node (P, V, S Series) systems, you will also have to option to go through a node shutdown.
In addition to selecting the node, you can also select a drive to get the information on the drive itself as well as have the ability to work through a drive replacement wizard. While this task, along with disk add, could be done through vCenter, the engineering teams between Dell EMC and VMware have worked to make this workflow easier by putting it into a wizard.
The other view within the health tab is “logical”.
As you can see here, this provides more of a “logical” view into the system. It provides some realtime data with regards to performance info (Storage IOPs, CPU Usage, Memory Usage) as well as capacity utilization information. In the bottom half of the screen, you’ll also see logical view into the nodes and their components. One thing to note here is the ESXi boot disk. That device is either a SATADOM (Quanta and 13G nodes) or a BOSS Device (14G nodes). VxRail actually doesn’t boot off the drives you see in the physical view – those are all allocated for vSAN storage.
The last tab is the “Config” tab. This tab has 3 different views we can look at: “System”, “Market” and “General”.
The “System” tab will show you the version of VxRail software running on your cluster. It’ll also point out what versions is most recent (in this case it was 4.5.225). You can choose to upgrade your cluster to that version via a “local upgrade” or “internet upgrade”. More information on the upgrade process is HERE.
After the “System” tab is the “Market” tab:
The market contains a few software options you can add to your VxRail cluster. The software has been tested on VxRail and in some cases, there may be some licensing included (like RecoverPoint for Virtual Machines). I like the market because it provides end users with the option to download and/or install some value add software to their cluster for cloud storage, backup and recovery or replication.
The last tab is the “General” tab:
This view will provide you with quite a bit of information about the cluster, but also with the ability to on/off some functionality. First, you’ll see some general information about the cluster software version and the ability to set a support account (this will help you open cases and search the KBs in the support tab). One other thing that’s nice is the ability to general a log bundle. Given that VxRail is more than just a hardware appliance, this button allows the system to get the logs from all the components in the cluster to help improve troubleshooting and reducing the amount of time to resolve open cases. You can also setup ESRS (Dial home) or a proxy if you need to connect to the outside world via a proxy server.
In the 2nd half of the “general” view, you can tell the VxRail system it’s “disconnected” from the internet to help disable the features that require connectivity outside of your environment. You can also “mute” cluster health monitoring. That feature is useful if you’re performing maintenance on the system. (Side note – this is “muted” during node add and cluster upgrade.) The last thing you can do is shutdown the cluster and change the language.
I know some people may pause at the “Shut Down” button. Have no fear, this isn’t a single click and your cluster is down. After you click the shut down button, there’s another pop-up window that asks “are you sure?”. Then AFTER that, there’s a cluster shut down precheck:
If you notice, part of the precheck is that the system will “Check and ensure that all customer virtual machines have been shut down”. That basically means you have to FIRST power off all non-VxRail VMs before you can shut down a cluster. This helps avoid an “accidental” shut down because an administrator would have to accidentally power off all the VMs on the cluster first.
All in all, you can see that VxRail Manager is all about providing information as well as an easy way to administer the environment from a hardware and software life cycle management perspective. The day-to-day VMware activities are still carried out through vCenter. The interface is pretty easy to maneuver, and the future is only going to get easier – stay tuned!