This session will walk through deploying Cisco Nexus dashboard on Nutanix AHV, focusing on the preparation and network requirements.00:07Here's the path we will follow. 1st, we'll discuss the AHD platform prerequisites and readiness. Then we'll begin the deployment. The demo that follows is a single node deployment. The process to installing node took approximately 60 min.00:18If you are deploying a three node cluster, the individual node deployments are the same. The difference will be when you add 2nd and 3rd nodes to the initial node in the Nexus dashboard GUI to create the three node cluster. Before deploying.00:35Nexus dashboard on AHV confirmed that the AHV release and available resources meet Cisco's current deployment guidance. Download the Nexus dashboard 4.3 .1 Qcal two software bundles from Cisco. Next Nutanix, AHV is supported as of Nexus dashboard 4.3 point.00:51.1. The image is approximately 15 gb. You will need to extract it, which will require double the space requirements. I'd suggest starting with 60 gb of free space. You will need to deploy a Linux VM in the AHV environment for the deployment.01:11The deployment script runs on Linux. You can use that Linux VM to validate the IL latency of the Nutanix storage device. For reference I'm using an Umbutu 26 oh four.one VM. No version is specified in the deployment guide.01:29As long as the VM can provide simple tools like TAR, VI or nano, and Bash. Gather required networking details to complete the installation. This will include Nutanix details, virtual switch information, and credential information. I will show a table of the required information later.01:45For sources, please look up Cisco Nexus dashboard deployment Guide on your web browser. I'll show the software download and documentation links shortly. A configuration and deployment script is supplied as part of the image bundle.02:04This will assist with creating the YAML configuration and deployment file. Be warned there is very little error checking creating the configuration file. Error checking will be performed when the deployment begins. If you make an error in the configuration file, it may take several minutes before the error results in a deployment failure.02:20You'll need to correct the error and start over. You will use the same script plus configuration file and Q out two image deploy a Nexus dashboard node. You can monitor the initial deployment progress in the VM console. Once this completes, you will be provided with instructions in the VM console to access the.02:39Nexus dashboard GUI to continue the deployment. At this point, the deployment is the same as on any other hypervisor or appliance. Answer the configuration questions to complete the node configuration. After the node or cluster deployment completes, we will validate the installation.02:59We'll start with software.cisco.com and go ahead and access downloads, and then search for Nexus dashboard. We wanna start with version 4.3 .1. This is the 1st version that supports Nutanix A HV.03:23We will download the Linux KVM bundle and place this on the Linux VM that's on HV. Then go to the release notes. Please read the release notes. Then we can go on to related content.03:42This will give us a link to all things Nexus dashboard. From here you'll see all the information you probably ever need for Nexus dashboard. I'm going, we're gonna go to deployment and upgrade. Make sure you're on 4.3 .1 or four point three.X.03:58And then go to deployment overview and requirements. Here you'll see deploying the cluster, installation checklist and various different platforms. Let's start with a pre installation checklist.04:17It will help you guide you through a creating a table of the information you'll need to deploy Nexus dashboard cluster successfully. It's quite a bit of information, but it is all required. NTP DNS, information about management IP addresses.04:35Data network and if you've got multiple nodes, additional information for those nodes. Then we can go to deploying a virtual Nexus dashboard in Nutanix. Again, you've gone through the prerequisites, pre installation checklist.04:52Some version information. The next thing we want to do is on that Linux VM, we're going to verify the I/O latency of the Nutanix storage device. On that VM, you will install an FIO package and then run a script that will run.05:10And test the I/O of this storage device. I'll show this in the demo, but you're looking for a specific value in this range. Make sure you understand what you're deploying. Is that a data node, an app node, and so forth. Each one of them will will have different storage requirements.05:27Port information to access Nutanix. The bundle will need to be extracted, and once it's extracted, it will create a KVM developers deployer script and the software image itself.05:50The KVM deployer will help you create a configuration file, and it will walk you through answering all the questions to create the files successfully. The cluster size, what VM type it is going to be.06:06Information about Nutanix and the Nexus dashboard itself. You can use the same script to create a three node cluster, again, just answer all the questions. From there, we'll use the, you can actually cat out the file or VI it for simple changes. If there's anything elaborate I would recommend just rerun.06:21Running the script and reinputting the data.06:41Once you've created the configuration file, you will use that configuration file and the KVM deployer script to deploy onto Nutanix. It will start showing you all the installation steps. You can also bring up the VM console and watch the initial configuration.06:44Once the initial configuration is completed, the script will show you or the console will show you how to access the Nexus dashboard GUI to complete the installation. This is my Linux VM on the Nutanix HV. I'm going to run the FIO package.07:02With the options in order to test and validate the I/O status of the Nutanix storage. The 1st thing I'm gonna do is download the FIO package. Now this has already been installed on this Linux VM, so it's going to be a NO op, and they also recommend making a test directory for.07:22Storing the results.07:42And then I'm going to run the FIO package with the options as specified in the documentation to validate the I/O. Again, this is usually fairly quick. Once it reaches 100 %.07:46What I'm looking for is this F sync sync file range field, and I'm looking for this 99.99 percentile field. And this is in microseconds, so this is approximately 5000 μs or 5000 milliseconds.08:07The requirement is less than 20 milliseconds, so we meet that. I just want to point out on our Nutanix infrastructure in our lab, we have a virtual switch VS zero, and in that.08:24We have several subnets defined. The ones we are interested in is this native 1096. This is our out of band network, and this is what we'll use for the management of the nexus dashboard. And we also have a subnet defined as VLAN 2096 and the data. This will be our fabric, our data.08:40Network for Nexus dashboard. And again, just to try to illustrate the point, our Nexus dashboard virtual machine makes use of two networks, the VLAN 1096 network and the VLAN 2906.08:59And the data network, and this is how kind of all cooks together inside the Nexus dashboard cluster Nutanix cluster. Again, this is in the V switch area of next of Nutanix, and again just trying to make sure we understand what goes where.09:14And then again I create a table that has this information along with whatever else was required in the pre prerequisites and to configure the Nexus dashboard inside of Nutanix. I'm back on my Nutanix VM and I've also got a file with all the.09:32Information I need to fill out the configuration. I've got my KVM deployer script and the Qcal image. So I'm gonna run the KVM deployer and again it's just gonna walk me through the information it needs to do.09:52To create the file for the configuration. So this is gonna be a Nutanix, and the image path is going to be where I am, and a single cluster, a single node in the cluster.10:12And I'm gonna make it a data node, and this is the admin password for the Nexus dashboard admin or rescue user and then the Prism central IP address.10:31The Nutanix, this is the username that's creating the VM, and the password i went for that, that's on Nutanix.10:51Yeah, and then a cluster name, this is user definable. This is the V switch name for the management network.11:09And the data network, number of CPUs default two. I'm gonna make this node name, this is what we'll see in.11:28Nutanix VM, the IP address of the management network or the out of band network, Management gateway, and the name of the YAML file when we're done, so you can review it.11:44Here, again you can either proceed or continue or you can go ahead and continue and edit it. It does hide the passwords. Again, it's good idea just to review it and this is going to be my name that I give the configuration file.12:04I can't cat out this file because it will display the hidden passwords but, this, you can edit it if you want to for simple simple changes. But again, if you have any issues, just rerun the script. It's fairly quite easy. At this point, you can deploy Nexus dashboard using a configuration script.12:23And it'll start going to town on it. It's now deployed, I can go to Prism and continue with the console.12:43And Nexus dashboard is beaking its boot up process.13:18At this point, you can continue to the URL specified. This will get you the management GUI of the nexus dashboard. You'll enter the password that you put in the configuration file.13:34Once you've logged in, you're gonna go to the cluster bring up.13:57I'm gonna use my cheat sheet to get this information more or less correct without too many typos. I'm gonna do a LAN install, the SAN install is almost the same except the SAN install will allow you to use one subnet as opposed to two.14:00I do need to provide a DNS entry and NTP.14:20Proxy is optional. Our lab does not require a proxy to get to the outside world. You may need to provide this if you have a restriction. The proxy servers used to get to things like intersight. If you go ahead and skip it. It will complain, it will warn me that it.14:41I make sure I understand why I'm skipping the proxy and some advanced settings. These are networks that are used by the containers to talk to Nexus dashboard. If you do need to change them, please review the configuration guide.14:56Now I need to complete the data network configuration. I need to provide IP addresses for the data network. And I've got a node name for the data network.15:26And IP addresses. So this is the data network IP address, mask, and the gateway. If I had.15:41Multiple nodes for my cluster, this is where I would add node two and add node three. In my case I'm just doing a single node.16:02And again, it's gonna let me know that, hey, normally you would do three nodes for a production environment, but this just gonna be for a lab. I'm gonna go ahead and continue. I do need to add three persistent IP addresses, and this is for.16:10Everything to talk to the data pool that's being constructed. So this is gonna be my 1st IP address. This will be on the data network, and I'm gonna add the.16:26The other two persistent IPs, Continue.16:45It's gonna give me an opportunity to review, make changes if necessary, otherwise, I'm gonna save and continue. The installation will continue. It's gonna take probably 30 to 45 min. I'll I'll put a skip notice as I move ahead to the final installation.17:02Yeah During the cluster bring up, you'll see some status. If you want to see some things going on with the Kubernetes, you can SSH as rescue user to the CLI.17:21And you can run some Kubernetes commands.17:49And you can see the containers getting installed or status and so forth. And I can, you can pipe it to word count and just see the status. That's at 55. It'll be over 100 when the.18:05Deployment is complete, at this point, Nexus dashboards completed. It may not be completely 100 % active, but it is very, very close at this point.18:21We can look at admin system status and there's a few items still matching. So again, it's not completely running, but it's getting very very close. Nexus dashboard connection to Intersight is not there because we haven't configured it.18:41Additionally the backup has not been configured at this point in time, but like I said it's very close to being complete and we're up to a hundred and twelve services.18:58So it should be just a few more minutes before it's complete. We can take a look at some of these system settings. Again, this is what we've put in, so ideally it should be a correct the persistent IPs that we added here are being used by.19:16Collector services, syslog, and SSH services. But at this point, within a few minutes we can start adding fabrics and begin using Nexus dashboard.19:35Okay. I think we're just about there. Yep, everything's completely up and running except for backups and inter site. That concludes this demonstration. Thank you.19:50
