 Hi, this is your host in Bhartiya and welcome to another brand new episode of TFR Let's see or TFR demo and today we have two guests from cloud Casa by catalogic It's chief operating officer Satya Sankaran and software engineer MD slum Satya MD is great to have you folks on the show Thanks, I'll always a pleasure to speak to you. Thank you. Happy to be here. Satya Of course, we earlier talked when you folks made a switch to Valero Talk a bit about how is this transition working out for not only you folks But also your customers and the users who are leveraging it. It's been a fantastic ride in We went to embrace open source primarily because that's really where every single user of Kubernetes is starting out Right. Everyone's journey is starting with open source technologies first They use Valero and over a period of time Maybe they identify certain things that doesn't just work with Valero and they decide to go to commercial Solutions right but everyone's journey starts with open source even a competitors that knowledge that in that place And but what we saw is and you really look at market adoption You know, you've got Valero with 99% of the market share and everybody else fighting for the one person That is, you know switching over from Valero switching over from open source because well, it doesn't meet all my requirements And I can't wait for Valero to evolve to what I need In a very short time frame. So that's why we really said hey We want to embrace the open source and be where the customers are and the users are and we want to give them a Path to user solution without actually having to switch away from open source because you know It's almost a religious decision for a lot of people, right? You don't want to move from open source to close source if you can avoid it And and and that's not a conversation. We want to be part of it. So we won't keep those guys there So we feel very welcomed by this community. We started contributing very very early on in this decision We now actually answer, you know, more than 90% of the questions that come up in the Valero slack channel or the forums You know, most of the questions are answered by our team. And again, it's because That was what is needed in this ecosystem because the maintainers were a very strong developers But they didn't have a background of you know support and and and in supporting enterprise customers and so on So we took on that role of again kind of just owning that community support In the back end we will also started making, you know, buck fixes and and your back porting important fixes that aren't there in the Loadest version or the version that is most applied So we've started doing that and we're now actually working on some of the core features that are coming up To support cube water or features pretty big. We're working with it at maintainers We're working with the VMware maintainers to make sure, you know, those pieces are In place as well. So in a nutshell, we feel welcomed by the community. We feel like we're contributing to the community and That's that only makes the community stronger and we feel that for the users of Valero You know, we are a low-friction path to enterprise features without actually having to move away from Valero completely running Valero and still getting the features that you as an enterprise user will Will be looking for if I'm not wrong AWS is one of the first cloud providers that you folks Integrated with can you talk a bit about of course, you know, it's a great mix there AWS and cloud Casa The benefits that users see of using cloud Casa with AWS look, I think There are a dozen solutions out there that support Kubernetes for backups today How do you stand out on this ecosystem, right? We're the first ones that really fully integrated with the AWS EKS service And from day one when we supported EKS, we did it in a way that we integrate directly with the EKS engine itself There is so many and so much in Kubernetes that is stored inside Kubernetes But there are a lot of things configurations that are stored at the envelope layer of Kubernetes cluster itself, right? So that information is not sitting inside the cluster it's actually sitting inside the guy that's hosting the cluster and that in this case the most popular Hosting provided in this case is the KS, right? So we integrate directly with the KS to collect information on your node groups your VPC is your your your security group settings and And and MD will walk through some of the load balancer work that we're we did as well We're collecting all of this information In order to make sure that when you bring back your environment, you're not just bringing back the the cluster information that is sitting inside but also the cluster itself Right, so that's one of the benefits that you know users of cloud class I get with the EKS that you don't get to any other solution that is out there in the market second EKS itself is a is is recommending Valero as the best practice for doing backups with their ecosystem EKS is different in the sense that AKS now have AKS backups GKE now have GKE backups whereas AWS simply endorses Valero as the best practice option for for you and and We are the only real solution out there. That's fully compatible with running an upstream version of of of Valero You run the latest version of Valero or the version that works best for you And we simply support, you know management and control of that On top of EKS, right and the third thing an EKS customer should Consider when they are working with us is I talked about how we deliver agentless backups with EKS I talked about how we can spin up a whole cluster With EKS, but what we what a lot of people don't recognize is You can actually back up an on-prem cluster or a cluster that is running in an AKS or GKE and Utilize a recovery workflows to spin it up in EKS in the region of your choice and your account of your choice, right? So as a EKS user you get agentless experience. You get open-source compatibility you get full stack recovery and You get to use EKS as a migration platform as well If you want to move to EKS, you can kind of go from anywhere and we deliver that cluster and cloud mobility for those customers and in and all of those is a very Intentional integration into that is EKS engine because again, it's been the first one We started supporting and it's still going strong. Can you also talk about what does AWS offer for container backup today? Look, AWS has something called AWS backups. It actually supports five or six workloads But EKS is not one of them Kubernetes is not one of them and in in in in this context EKS or AWS is different from the other cloud providers because Azure has AKS backup that just went public preview not too long ago GKE added GKE backup. I think that went Public preview or G. I don't recall but sometime late last year. So now the other two cloud providers offer a bit of backup built into their cloud provider services whereas AWS's Recommendation at this point and the best practice is to simply rely on the open-source Valero to do its backups, right? So AWS doesn't offer an EKS backup per se, but they recommend Valero This is where our strategy of embracing Valero and offering all of the things that AWS user is is is used to Multicluster management You know enterprise reporting and control from a central console Look, if you're gonna put all of your primary workloads and rely on AWS to manage that for you Why do you think people are gonna just do a DIY on backups? Right if primary workloads are going to be something that you rely on AWS to do backups are not something that you're gonna DIY yourself and There cloud cluster becomes an option while it is using the recommended Valero that that AWS recommends that creates a whole bunch of reference architectures in but it gives you that hosted experience a SAS provider offers And that's how we help AWS fill that gap With cloud awesome. Can you also talk a bit about? Why predicting and restoring network and of course load balance or configurations are Important in cloud because mostly cloud is seen as a magical place Because you can't just take a cluster from one place and try to restore the Kubernetes cluster in another place and expect everything to magically work Right people talk about the fact that the platform offers a lot of portability and mobility That just tells you that the binaries will come up and run It doesn't mean all the dependent services that it relies on Are going to work by simply taking a cluster from one cloud provider and try to run it in a different cloud provider, right? and in this is where How you build the envelope around your cluster is very very important. What services are you recommend or are you dependent on your cloud provider to give you? you know almost every AWS user has some level of Network connectivity that is again very proprietary to what AWS offers and there is load balancer that you often will use the default load balancer that is out there That is being offered by the service security groups almost every AWS user uses a concept of IAM Accounts and enable that certain people have certain certain roles and access controls and so on in all of this information Needs to be captured in some way So that when you bring back a cluster you can either map it map this to a different account map this to a different region or potentially even to a different cloud provider, right so It is important that you bring your whole self When you bring up a workload you bring up the whole workload not just the underlying Binary is in the container images in the cluster configuration, but also how is that cluster, you know, what is the cluster dependent on? Can you bring back all of that? During the restore process as well. That's what we make the job easier Otherwise the alternative is you know someone has to manually go and configure everything make sure that these settings match the settings that were taken at the time of backup Which is great on paper. Yeah, a lot of people can do that But oftentimes you don't even know what worked once you've lost the data, right? If you don't have a great recording system, you're not even going to know how it was previously configured If you don't record that information during backup And so we fill all of those gaps so that you don't have to rely on a manual process to get a cluster up and you don't have to You know sit and record and transform when we can do all of that for you automated and essentially deliver recovery is cool MD, that's your turn. Let's do the demo though. So I've logged into my cloud Casa account here and So this is the cloud Casa dashboard So what I'm going to be showing today is I'm going to be doing a backup of an EKS cluster Along with all the associated AWS resources like the VPC subnets The load balancer Satya mentioned And then I'm going to be taking that backup and then I'm going I'm going to be doing a restore from that backup So the first thing that you need to do is to actually Register your AWS account with cloud Casa So to do that you would go to configuration cloud accounts And then you click on add cloud account button on the right side here, and then you can select AWS as the provider type and click next And then you see here you have this launch stack button which will open up the cloud formation stack page on AWS and here you can use our cloud Casa cloud formation stack template to actually Register your AWS account and then you can see here there are specific permissions Specific features that you can enable that corresponds to specific permissions on on AWS rights like EKS, RDS and EBS volume stuff like that But in my case I've already registered this AWS account So I can show you that here As you can see I have one AWS account registered In cloud Casa And once you register your AWS account, it'll be an active state So once you do that you can actually click on the AWS account And then you can see that cloud Casa has discovered All your AWS resources and in this case You can see that I have two EKS clusters that were discovered And so you can see here one is in discovered state one is in an active state So once you register your AWS account of all of the clusters will be in discovered state But to get it into active state you actually need to install our cloud Casa agent And to do that you go to action and install and then you can see here You have a keep control command that you can copy and paste to actually install our agent So the the EKS cluster that I'm going to be backing up today is this EKS cluster here md-test-2 So if I can show you on the AWS console right so over here I have two EKS clusters And this is the cluster that I'm going to Back up here So I'm going to copy that and I'm going to go to my terminal here I'm going to connect to this Cluster here Right it's the cluster here Okay, so you can see here that in the cloud Casa IO namespace I have the cloud Casa agent installed And then you can see so I have a few web applications running here In three different namespaces the web one web two web three. I have some WordPress sites running And I have so if you look at See the persistent volumes Right you can see here that I have six volumes and I have so let's look at the Load balancer setup here. You can see here that I have the AWS load balancer controller running and I have So let's see here Right, so I have a service This service here, which is a load balancer type. So this is my first load balancer here and see if I have any more Right, so I have an ingress running. So that's my second load balancer here And then if I look at my target group bindings Right, so I have you can see here. So this This is the service. This is the ingress And then this is a third load balancer that I have here, which I created manually That I can show you on aws So over here if I go to the load balancer page here You can see that I have three load balancers associated with its eks cluster So two network load balancers one application load balancer And one of these load balancers as you can see is Was provisioned outside the scope of the adb load balancer controller So this one could have been created manually or through something like terraform So what I'm going to do is I want to back up everything right? So I want to back up my eks cluster the configuration the Kubernetes resources the persistent volumes and the load balancers vpcs. So I want to back up all of this information right So to do that from cloud casa what you would need to do is you would go to protection backups Then you can define a backup here And select your cluster Click next and then you can so we have options to select specific resources But I'll leave everything at the default here. You can see here like full cluster. So you can select namespaces select resource types labels Include persistent volumes or this also you can do a snapshot only or you can do a snapshot and copy So you can actually copy the data To cloud casa storage or you can specify your own storage But I'll leave everything here as it is like that and click next So there's hooks pre post a free backup hooks post backup hooks. I'll leave that as it is and click next So there's policy information here You can back up based on a specific schedule that you provide but I'll leave that as it is And then here you can just set up a you can just put a backup name Or you can call it anything like demo one or whatever And then you can click create here So I won't be doing that now because I already have a backup that I ran yesterday So I can show you that so if I got a protection backups And I click on this backup here. You can see I have a backup that ran yesterday Um That is completed and you can see that all my persistent volumes were snapshotted and copied And then if you look at the activity log, okay, so here you can actually see what was backed up So if you look if I scroll down here, you can see that So it backed up all the kubernetes resources, right? So the namespaces The pvcs And then you can see here it backed up the configuration of the eks cluster The node groups the vpc the subnets the security groups And then you can see here that it it did back up that one non-controller provisioned adbs load balancer And then if you scroll down, you can see that it also copied the data as well So yeah, and then the ingress and service and target group bindings as well as you can see So that's for the backup. So once you have that backup You can actually restore From that backup. So if I go to restores, I can click on define restore here Or I can go to recovery points and I can select the latest recovery point over here and click on restore And just like backup you see you can select specific resources or resources to restore like select namespaces or labels or resource types But I'll leave everything as it is and click next Okay, so here you can Select the destination So you can either restore to the existing cluster So the md test to cluster or if you have another cluster registered you can restore to that cluster Or in my case, I want to create a new eks cluster Based on the configurations in the backup, right? So I click on that and then click on eks cluster And then you can select an adbs account which account you want to restore to so in my case only have one account But if you have multiple accounts, you can actually restore to a different adbs account So for here, I'll click next And you can specify name for the eks cluster. So like demo one next The adbs user. So this is the user that will have access to this eks cluster once it's restored So I have that Set that to my user that you can select the region what region you want to restore the eks cluster to And the other field these are all I'll leave it as it is these fields also exist in the eks adbs console. So I'll leave that as it is But all of these informations are pre-filled for you from the backup So it's getting all this information like this, for example, this network information Is coming from the backup. So I can actually have cloud costs that create vpcs and subnets for me From the based on the configuration it got in the backup Or I can Have the option of specifying what vpc and subnets I want to use But I'll leave this as it is and click next Same with the notable configuration. So all of this information comes from the backup Or you can customize it as you want, but I'll leave it as it is and click next Okay, so this is the page that we wanted to come to. So this is the load balancer section So here I can actually So it's asking for the the I am role for the controller the load balancer controller. So I can actually specify What what the I am role here is so I can do that right? So this has permissions to access the The load balancer on AWS Um, and then over here, we can actually modify certain resources the kubernetes resources Uh, so for example, like the service if I want to modify the service, right? So let's say I want to modify that uh, wordpress service that you saw earlier, right? Let's say this wordpress service is the load balancer type But let's say I want to I don't want this as a load balancer, right? I don't want to expose the service I can just set up a cluster ip and save it here or I can modify the ingress for example, right? Let's say this ingress resource here. Um, let's see Let's say I want to change the host, right? So I can change the host here Right now I want to change the host I can do that and I can of course I can also do target group bindings I can update this as well But uh, okay, and uh, so the last thing here is the uh, the non-controller provision load balancer So this is the uh, the non-controller provision load balancer that I mentioned before Right? So this could be created manually or through terraform Um, so this one is coming from the backup. You can choose to have quadcastle restore this for you from the backup Um, so I'll leave that as it is click next And these are the list of add-ons, um I'll leave this as it is so the uh, the uh, the enroll for the ebscsi driver. I'll leave that as it is Uh, so you have the options here. So rename namespaces change storage classes preserve node ports If you want to um, or an overwrite existing resources if there are any Click next post restore hooks. I'll leave that as it is and over here. You can just set a name Right? Uh demo restore, right? And then you would click on restore Here So I already had a restore that I ran yesterday So I can show you that here. So if I got a protection restores and I click on The restore that I had you can see here. I had a successful restore job that ran yesterday Um, and then over here if you look at the pv details, you can see that all the The data of my person volumes in the in the backup was copied over to my new eks cluster And you can see here. So what what did it do in the activity log, right? You can see that It started the restore of the eks cluster. It created the vpcs and some that's from the backup It created the new eks cluster in the in this region the us east one It created the node pool node groups And then it gave access to my user. Uh, and then it You know, it set up the oidc provider. It set up this service account with the EBS CSI driver and then you can see here that you created the the The non non aws load master controller provision load bouncer, right? So the one that was that was created manually So it created that And then it set up the adbs load bouncer controller here as you can see Uh, and then you can see here that it restored all the kubernetes resources including these services and the in the ingress and all that stuff So all of that is now restored. Um, so basically what what just happened is all of the the the configurations of the eks cluster and uh, the kubernetes data the kubernetes resources the personal volume the subnets security groups the The the vpcs the load bouncer as all of that has been restored from the backup Satya md. Thank you so much for of course talking about velaro cloud cosi and thanks for this demo And I would love to chat with you again. Thank you. Awesome. So now and always a pleasure talking to you. Thank you