How to install Proxmox, cluster & setup High Availability on multiple host nodes!

we're going to install proxmox as a hypervisor on all these servers and then cluster them all together why because we can that something you want to see stick around let's get to it [Music] what up Geek Squad so today of course talking about virtualization the hypervisor choice for this video is proxbox what is proxmox and why would I want to use it over something else that's an excellent question I'm glad you asked so here we are on uh the proxmox website main page today we are going to focus on virtualization proxmox virtual environment or a hypervisor a hypervisor is a very small operating system that lives on the bare metal of your server and allows you to allocate those bare metal resources in a virtual manner proxmox is an open source server management platform for Enterprise virtualization with the integrated web browser user interface you can manage VMS and containers High availability for clusters which is what we'll be doing today and integrated Disaster Recovery tools with ease while many people start with a single note proxmox can scale out to a large set of clustered nodes so getting started you'll download the iso image create a bootable USB or CD and then configure the host machines so where do we go from here well I'm going to Trek my butt out there in the cold install the hypervisor then we're going to connect to them and we're going to set them all up in a cluster spin up some VMS migrate live VMS from one machine to another and uh and we'll go from there a few moments later here we are middle of November out in the big cave uh freezing off the took us all for you what we're gonna have to do first is uh create a virtual drive we'll end up erasing what's already there creating a new Virtual Drive in a raid configuration or non-raid however we want to do it so that we have something that we can install the hypervisor onto right here we go here we are all right so in our RAID controller can clear the foreign configuration so blew away the old virtual disk Frick gets cold go to Hardware configuration do a raid five now we're at the point where we're gonna plug in our Virtual Drive 328 a.m at one time bios boot menu optical drive connected to front USB virtual CD-ROM right so we got our proximox menu go to install proxbox so here we can select our different drives see we can see this is an external hard drive which is what we got in here but we're obviously going to pick our Virtual Drive which is built in the physical host and here we go 12 seconds later so here we are we've got our first uh proxmox node uh PVE and uh the other tab you can't see it up here but the other tab is pve2 so now comes the hard part I'd say that sarcastically because it's not difficult at all so we're gonna go into what we want as our masternode really you can do this as any nodes because you can manage them from any nodes so we go into Data Center on our on our node go into cluster you can see we don't have a any clusters here yet so we are going to create a cluster and we'll name it Geek Squad no spaces no underscores just numerics I guess so here's our cluster now we can go here to pve2 under data center cluster join cluster what you're going to put here you're going to get from the join information under cluster info in the new cluster that you just created so we'll hit copy information come back over here and paste it put in the root password join Geek Squad which reminds me you should subscribe to the channel and join The Geek Squad now we can over here we can see that uh under data center on the first node we've got pve1 and pve2 here uh so we've got both of our nodes set up we can expand the resources and see that we've got some VMS over here so what have we done we've spun up the hypervisor on each of the physical hosts we've created a cluster we've joined a second host to that cluster uh now let's power on one of our our VMS here's how easy it is to move a VM from one physical host to another in proxbox so we've got our list of VMS here we can see that Windows 11 is is running it's got the little green play button here a little green triangle so we've got our VM here that is playing at the same time which means streaming audio connected to the internet doing its thing we're going to right click on the VM and come down here to migrate and choose which Target node in this case we just have the one the pbe2 and all we got to do is hit migrate now this is supposed to be a a live migration shouldn't interrupt it should just keep going we can see that oh my gosh that is going to take a very long time if we had a shared storage node that was hosting the storage layer of the VM it would be super quick this is going to take a very long time so I'll be back [Music] okay so I kept recording because I knew it was I thought it was actually going to take a lot longer than it did I'm only using a one gig connection between the servers so we moved to VM from one host to another without the end user noticing any difference uh that's pretty sweet we can see that we've got our VM over here on uh on pve2 that was easy day two so we looked at how to set up your redundancy on the hardware level with raid and a RAID controller in the server what we're going to do now is create non-raid disks and our redundancy level is going to move to the virtual layer this will allow us to set up a ZFS pool and set up high availability between three physical hosts so let's check that out don't start this if you're in a hurry and here we are back at the life cycle controller what we're going to want to do is go all the way down to system setup Advanced Hardware configuration from here we'll go to device settings we'll go into our RAID controller card controller management and we're going to scroll all the way down to Advanced controller management then the very last option is switch to HBA mode takes a reboot to take effect we can then exit all the way out physically restart the host it should then start in HBA mode non-raid setup so each drive would be recognized as its own separate drive here we are in the boot manager we're just going to go down to one shot bios boot menu from here we've got different options of where we can boot from we're going to go down to which one is it uh optical drive connected to front USB 2 virtual CD-ROM all right here we are install proxbox just like previously this time we're going to see three different drive options currently because of what I have available three 300 gig 10K drives in each system you could use a smaller drive a 64 32 but remember if you enable logging those logs are going to be stored on your boot disk we've got our discs a b and c as it's recognized here on the installer we're gonna pick a but they're all three the same size and we are going to actually name this ha1 management interface this is not going to be the idrac interface IP address we'll do these in the 210 schemes 2 10 11 and 12. I don't have a DNS server here so we're just going to use Google's and we're going to hit install so it's going to go through and install the hypervisor on the OneDrive which is going to free up the other two drives as Standalone drives so from there we can create a virtual raid and a ZFS pool that then will allow us to create a high availability cluster we're going to let this install I'm going to install this on the other two nodes and I will see you back at the command center several bad puns later here we are back at the command center we've got our three hosts ha one two and three I clustered them together using the same process we just looked at a little bit ago now we need to create a ZFS storage pool that all three of the nodes have so that the VMS know where to look when they're migrating and moving from one physical host to another so first things first we are going to click the first host here we'll look at our disks here we are we see that they're both here boom and boom now we're gonna go to ZFS create ZFS zfs-gs for Geek Squad and we're going to enable both of them on the first node you do this to you have to make sure that ad storage is checked the other two nodes you're not going to do this now we're going to raid these as well so we're only going to have a 300 gig ZFS pool but we'll have the redundancy on the software side and you can see here that ZFS is not compatible with drives backed by Hardware click create it takes a little bit of time now we can see under node one we've got our single ZFS pool and we're going to repeat this process for nodes two and three now we need to remember the name is case sensitive and it has to be exact to the first name that you created exact settings all of that and we're going to uncheck our ad storage box right here so create and we're going to do the same thing on number three there we go we have our ZFS pool on each node so now we're going to go here to Data Center storage for the storage menu we can locate the ZFS storage pool here I'm going to click it and we're going to add nodes two and three and however many nodes you have and we hit ok now we can see we're going to take a second here let this propagate and the status will change from unknown to available yep so it's available on all three nodes we can see these are not in high availability yet each node can still control its own storage so now we have to set up the replication in order to replicate we need to have something to replicate so we're going to spin up a simple VM something quick and easy all right so this video isn't going to get too deep about how to spin up VMS manage and control them really but to just for quick reference within one of the nodes we'll hit upload under ISO images under storage so let's go back probably click storage and it'll just so summary here click on local go to ISO images so if you come down here to lvm you see that you don't have the option to put ISO images so just under ha1 we're going to do ISO images upload select a file file name Ubuntu 22 upload and now we wait so I don't have one gig connection from my local machine here all the way out to the data center it's only uh 100 Meg but we'll change that in a future video we've got our VM spun up just did Ubuntu it's quick and easy we'll click on our VM we'll come down to replication and we're going to add a new replication job by clicking add Target is going to be ha2 for now we'll leave it as every 15 minutes we'll hit create so remember replication is a resource intensive task so if you've got a lot going on it can bog down your your physical host doing that replication task and we're going to add the same thing to node three this was already synced should be able to there we go schedule now so we click on uh on the on the job on the replication job hit schedule now go ahead and start the replication now we can set up h a this will ensure that the VM is running on one server at all times if the cluster detects that a node isn't available the VM will be migrated to another node and auto booted to its last replication point so we're going to come over here go to h a whoops in our data center menu uh under resources click add and a VM to monitor so the one VM that we have out there uh leave everything else here the same now we can see that the VM is being monitored and that's it so all right here we are we've got our VM on ha1 it's here running so we've got our continuous ping going on it's living here on ha1 we've got our console here that's funky Let's uh let's see we're recording and we're recording so we're gonna go pull the plug on ha1 in real time and three two one pulled the plug one thing we're gonna notice here this is the ha1 console so it's offline so it's not going to respond to anything now so we're going to move over here to ha2 we can see ha1 is offline we're still pinging so we're gonna see just um just how long it takes we can see that ha1 is offline oh and it looks like our VM is now live on uh on ha2 our ping started pinging again uh oh okay so VNC is not going to work because you can see it's looking at the 210 IP address and ha2 is to 11 but it boots up right to the state where it left off there we go that's exactly what I was trying to accomplish so essentially we could have a crypto node a flux node as endnote or whatever node that's running on an aha cluster and if something fails on the first one the whole note the whole machine goes down it'll automatically jump to the second node and start running again right where it left off so as long as we have the commands that will start the actual node at the OS level uh without you know touching anything um it'll start back up on its own as I understand it as long as it's not benchmarking at the time then it goes down it's not going to send you to the back of the line so to speak for your node payouts so let's recap here we installed proxmox on the physical hosts we clustered those physical hosts together we then did a live migration without losing access to the operating system the VM remained live as it migrated from one host to another manually we then spun all of this back up on a separate Hardware cluster for high availability stood up as ZFS storage pool then we enable High availability we then successfully simulated an outage on node number one where a live VM was then migrated and auto booted back to its last current state on node number two I'd say that's a success make sure you subscribe to the channel if this is the kind of content that you are interested in seeing we're going to do some more things uh when it comes to proxmox and server type environments we'll do a video on some shared storage do a video on actually spinning up VMS and creating VM templates uh maybe containers versus VMS what the difference is why you would want to use one over the other backups and snapshots uh and and much more there's lots of different home lab apps and systems that I'm going to play with and try out and uh hopefully you'll you'll get to come along with me that didn't come out right so leave a comment down below let me know what you think about these kind of videos this is something you're interested in hit subscribe hit that Thumbs Up Button help with the YouTube algorithm and of course thanks for watching

Channel: Geek Of All Trades

Views: 9,524

Keywords: geek life, geek squad, geek of all trades, proxmox, high availability cluster, server cluster, virtualization, proxmox 7, home lab, home server, vm migration, virtual machines, homelab tutorial, flux node ha, flux node server, high availability server, proxmox ha, zfs, zfs replication, setup zfs, zfs raid, server storage, storage clustering, server ha, proxmox cluster setup, proxmox cluster, proxmox ve cluster, proxmox tutorial, proxmox cluster 3 nodes, how to

Id: UETNQJAz3Io

Length: 18min 46sec (1126 seconds)

Published: Sun Nov 20 2022