 I'm here from Lawrence Systems and RAID is not a backup, but RAID does offer resiliency from things like hard-dry failures. So if you have a bunch of drives in your Synology NAS and one of them fails, you should have some parity setups. So there's data and redundancy across these and you should be good. But what if all the drives fail or the entire system fails? Well, this is where Synology offers a high availability solution. And it's really easy to set up. So I wanted to walk through a tutorial here. So I had a couple of DS3617 XS models that are identical, which is a prerequisite for doing the SoundGHA. They have a lot of different models, even some of their smaller ones that support this. It's not only a high-end feature, but it's a really cool feature to load the high availability application, synchronize the two servers, and then have all of your data, not just in one SoundGHA server, but in two of them. And it is important that these servers are identical. There are some notes in SoundGHA. There's ways to make it work if they're not, but it's not recommended. So if you are planning this out, you get two identical Synology servers, and it's the best way to start out. But I want to walk through how that works, how to configure it, what it looks like, and then unplug it. Just show how you take a server, fail it over and see what happens and show you how it recovers and how fast it can recover, even when you're using it as a storage target for something like virtualization. Before we dive into the details of this tutorial, if you want to learn more about me and my company, head over to LawrenceSystems.com. If you like to hire a shared project, there's a hires button right at the top. If you want to support this channel in other ways, there's affiliate links down below to get you deals and discounts and products and services we talk about on this channel. All right, let's start out here talking about how our lab is currently set up. We have one labeled alpha at 3.52, one labeled beta at 3.53. And then we have a switch. Now, each of these are connected to this switch at 10 gig and the XC PNG hypervisors also connected at 10 gig. There's an add on card in each of these units. They're absolutely identical, including the add in card, the Synology SFP two port connector. This is all connected with SFP cables keeps it all really clean and simple. Now the connection is only using one of the ports on here and one of the ports on here to connect to the switch. And what we're going to do to build our HA cluster is connect this and this directly together with a debt cable. Nothing in between. We don't need a switch for the heartbeat. This is an important aspect of the way that these clusters work to get the cluster working. We're going to load the Synology HA cluster tool. We'll walk through the wizard and then we'll assign a cluster IP address. These will keep their current IP addresses of 3.52 and 3.53 and 3d.54 will become our management address. What the management address and also the shared address between is the one that the active server is handling and whichever server is active, that's the one that handles it. And when you're setting up things like your hypervisor or any of your access to these particular servers, you'll always use that IP address. These will still have these IP addresses on here, but having the management address is so you know exactly that that address is the one that's always going to be whatever server is active. So they actually kind of pass that IP address back and forth. And that's what we'll do in the demo when we unplug the system. Now the heartbeat is actually two purposes. The DAC cable that's connecting in the extra port on the SFP card is facilitating the heartbeat so these two servers know where what they're doing it each time and synchronization. So when data comes in from any of the external servers over the line and connects to whichever server is active, and let's say that the alpha one is active, the data flows in and goes across the line. You don't want the data also flowing back across that line to synchronize all the changes going on to this secondary passive server. So the heartbeat has to be the same connection if not faster than the incoming connection for where the data is. If not, you'll end up with well an asymmetrical connection where the data comes in potentially faster than it can go out and it would not be able to stay in sync with each other to perform proper failover. So they do have to all be at the same speed when they are for this case, they're all at 10 gig. And there's not really a switch needed or really even recommended to go between the devices. Putting a DAC SFP cable directly between these devices is the easiest simplest way. You don't need other things, you know, interfering with it. You could run it through a switch theoretically, but I recommend doing it this way. And it's in the Synology documentation as well. Now let's go on to how to actually set it up. It's pretty easy. Now the first and obvious thing is probably understanding that you need Synology High Availability application installed. Pretty easy to install it. We're going to hit open it up. And now it's going to create a High Availability cluster. It wants to confirm that we have a cluster connection on the switch just like we did in our diagram and the heartbeat connection between them. Create High Availability cluster. Yep. Notice we need a couple of cables. We need to choose the cluster interface. And for that, we're going to go over here and look at the network interfaces. And there's a 3.52 on LAN 5. So the LAN 5 is that. If you notice, there's nothing on LAN 6 that says DHCP, but it can't find a DHCP server because it's actively plugged into the other system. It doesn't need any other settings because Synology can take care of that for us. So we're going to go over here and say that's the heartbeat interface. That's the direct connection, no switch plugged into the other Synology. We're going to go next. Now it sees the other system and we're going to put the username and password in for the other system so we can go in there. We don't have to actually log into the other system directly. We're logged in just to the first system. Next, now we're going to give it a name. We'll call it the Turbo Cluster because that's what my staff was calling it when we were doing the demo here. 683.54. This is just like the IP address that we had set over here, 3.54. So that's what we're going to use. Next, now it's going to verify all the requirements. Make sure that they're the same, make sure the network setup is right, make sure the heartbeat interface is talking. Then we click next and done. I understand all the data on the passive server will be permanently removed and cannot be recovered. I've already got something set up on the active server and it's going to start the synchronization process. It would take a little while because I've got an active ice cozy and storage and things like that is all configured on here. So this is going to take a minute while it verifies all the environment, gets all it's set up and then synchronizes the two. So they're in sync. This will take a little while. So we're going to use the magic of video editing to skip ahead. The high availability cluster is healthy. So the alpha one is at 3.52. The beta one is still at 3.53, but we're accessing everything at 3.54. Now, pretty simple. Once this is set up, if you need to switch something like which one's active or which one needs to be switched over, we can actually just hit switch over and say, yes, and it'll switch them over. It'll do so in a very graceful way. But before we go there, let's talk about the storage setup. Currently, as I said, I have the Synology set up with ice cozy. I have a ice cozy target. It's connected right now to a XC PNG server. And it says Synology HA right here. And of course, when we look at the host information, our target is going to be 54, because 54 is that IP address that is getting passed between the two servers, whichever one has that IP address is going to be the active server. So we connect this here. And actually, I got a Windows server up and running on it. So here's my Windows server 2019. And maybe that's something you want to keep up and running and not have to restart when the storage server has a failure. And this is where the HA comes in of note and not covered in this particular tutorial is I know I don't have redundant switches in this. Obviously, another layer would be to put redundant switches between here. Just of note, this does not protect from a switch failure. So FYI, I'm aware probably already comment because I didn't say it earlier in the video back to the video. The disc here being on the Synology iSCSI and of course, it's up and running. So we'll show you what happens when we do the switchover. We're going to close that and we'll do a graceful switchover. So we say switchover, we say yes, we get a warning that we know that things are connected like iSCSI. We're just going to throw caution to the wind here. Go ahead and let us switch over. All right, the other server now took over and it's now the active one. And what about our Windows? Is it break? Is it throwing hard drives errors in the event log or anything like that? Actually, no, it's perfectly fine. We can still launch things on it and open up and it doesn't really notice that something happened and it switched over to the other server. This is actually really handy. There's ways you can handle updates to the Synology system this way and, you know, everything will keep working perfectly fine. You can gracefully switch between them. You're doing it in a very controlled fashion. But of course, failures never happen in a controlled fashion. So before we get into the failure mode, let's talk about other services besides just iSCSI, such as maybe running Docker. So we're going to go over here and pull up the Docker and any application really that you load on here. And we have one container up and running up for one minute. And that's because we just switched over. It was actually running for a while on the other server. Anything that you have set up and configured on the Synology automatically replicates to the passive Synology. So the active one has all the settings in there. So because this was in the running state, it will restart. It actually doesn't do any type of live migration between these. The applications restart on the other server, but still all the data is in synchronization the whole time. So other than having restarted, the integrity of the application is there and the services are still up and running. So people may have been logged out if that ran something that would lose sessions upon restart, for example. But at least it's all there. All the other services such as active backup for business and other applications I have installed on here, they would all just be taking over on the other side with no problem at all. I have a couple of different random things that I have set up in here, but they're all just Passover. This does work for SMB as well, the high availability. So if you are using this for your standard SMB rights, maybe a window share with a bunch of users on there and the main server fails and goes over. Once again, they're none the wiser to it because windows will be only connecting to that 3.54 server. But now let's do that catastrophic failure and show you what happens. We're going to go ahead and just rip the plug out of the DS3617 and see what happens. We have our two Synology servers, the alpha beta and it's currently on beta. So we're just going to go ahead and rip that plug out and reach back here. Now, as I said, there's the SFP cards in here. So they're now not talking to each other anymore. It's directly connected. Well, can't really pan out easy enough to get that whole thing in there. But trust me, only one of these goes switch, the other one goes there and it's realized its friend is missing. So we're going to plug it back in as it takes a little bit to boot and we'll see what happens. All right, now let's go see on the screen the failover because it's starting back up. It'll take a few minutes. So we're getting the warning that unable to detect the passive server, it realizes it failed. So it just flipped over. Not a problem. This server is now active and it's telling us actually this passive server because it was active. So it just flips it around. But of course, the important question is, what happened over here? Did Windows die? Did Windows still work? Can we still look through folders and launch things? And yeah, works perfectly fine. No problem at all doing the switchover. The data is synchronized. Now, what's going to happen though? This is where things get a little bit interesting. When this server comes back online, they are going to have to re-sync because right now active data is going back and forth between here. Next, let's go back and open up our Docker image again. Ooh, that container didn't start up automatically on there. So fail on the container to start, but oh, future time here who edits these videos and checks for accuracy. And I realize I did not have the auto start enabled for this particular one. So it actually was supposed to auto start on failure on the other server. It didn't. I did forget to check the box. Just wanted to have this for accuracy in there. That when you do a failover, if you don't have auto start on a hard fail, it won't auto start. Simple as that. Wanted to get this clarification in there. Now I can just hit start and the container is running again. The more important thing for particularly this demo is that one, the data integrity is there so I can still boot things up. And I didn't lose my storage target on my virtualization server because, while keeping this server up and running, you know, might be kind of critical to things. Now, once it detects this, and we'll go ahead and fast forward a little bit, and it does to show you the synchronization, because obviously this has got new data that's been added to it from us reading and writing and windows. Matter of fact, here, oh, it happened pretty quick. So currently we have, we're at 99% it realizes the difference between them. So there's only 1% of data change. The longer you have it off, the longer it would take to change. Also, while the passive server is off, you still can do things such as add other applications, install things from the package center, it'll still let you do everything. And that's just that much more data it has to sync once these things are on there. So even while it's not in sync, it's loading up the package center and letting me go ahead and make changes. Go back over to like all packages. And let's say we wanted to open up something like install Synology office, install. And even though they're not in sync, it's pausing because it's letting it's doing so much synchronization in a background, but I can still say yes and finish this and it'll be one more thing that gets synchronized to here. It'll take a little while and you'll watch right here the transfer speeds between the heartbeat connection. Because as I said, it isn't just the heartbeat to the servers can determine that they're both working, but also so you can transfer that data back and forth and get them synchronized back up. Now until these two systems are completely in sync, it would be dangerous to fail over again because there's data missing on the passive server until it gets synchronized. Just kind of an FYI in there. So you don't want to try and pass them back and forth, but didn't take long. I did that real time there. So you can see your sync. Now a few other options here they have is where you can eject or shut down or restart each of these individually. Like I said for updates, this is where you'll change the network settings in case you want to change things, change some of the host information. These are services like the services you like to enable auto failover when the active server is unavailable. Now this is that auto failover as in make sure these absolutely run when there's a catastrophic failure. And so I SCSI, SMB, I don't have FTP checked. Hopefully you're not using that. NFS shares and web services. Then we have the storage where you can look at it here. And of course logs of what's going on, what happened. And you can actually facilitate this through the Synology Logging Server. So you can keep track of all of it. So hopefully you found this helpful and how to set up Synology HA. And like I said earlier, there's quite a few models that support this. I think it's a really cool feature they've added in, because well, there's not a lot of times HA is that easy to set up with devices, especially when you want to use them as virtualization storage targets. But they've kind of made it as simple as just buy a couple of Synologies that support it and run through a few settings, connect a switch to them. Obviously, if you really want to do this, well connect redundant switches where you take cards where one port goes to one switch, one port goes to the air switch, switch talk to each other so everything can talk and you build that level of redundancy. I'll leave links in a white paper where they explain that a little bit more in depth. If you really want to build this out to more enterprise level, Synology's got work instructions on it, but hey, the settings are the same. You just go through that wizard, like I said, and thanks to you know, you have an HA system. All right, I'll leave links to the things I talked about down below and thank you. And thank you for making it all the way to the end of this video. If you've enjoyed the content, please give us a thumbs up. If you would like to see more content from this channel, hit the subscribe button and the bell icon. If you'd like to hire a short project, head over to laurancesystems.com and click the hires button right at the top. To help this channel out in other ways, there's a join button here for YouTube and a Patreon page where your support is greatly appreciated. For deals, discounts, and offers, check out our affiliate links in the description of all of our videos, including a link to our shirt store where we have a wide variety of shirts that we sell and designs come out well, randomly, so check back frequently. And finally, our forums. Forums.LauranceSystems.com is where you can have a more in-depth discussion about this video and other tech topics covered on this channel. Thanks again for watching and look forward to hearing from you.