NetApp Astra Control protects Red Hat OpenShift workloads

[Music] [Applause] [Music] foreign [Music] do [Music] hmm [Music] [Music] so [Music] [Music] good morning good afternoon good evening and welcome to a special edition of red hat live streaming i am chris short host with the most of red hat live streaming i'm joined by two folks one from netapp one from red hat andrew sullivan red hat how are you doing yes sir i'm doing well the the host with the most the producer with the produciest producers producer you've ever seen i don't know i don't know where i was going with that anyways yeah it's good to see you a little strange to see you on a friday and not a wednesday but happy to be here and uh this one is is special um i i'm looking forward to this uh this is the first um partner live stream that i have been on um so i i'm excited i'm really happy to be here uh as you said we are joined by netapp's adam haid so adam if you don't mind introducing yourself good morning everybody how's it going uh my name is adam haye i'm a principal engineer for netapp um it's early in colorado uh it's eight o'clock i am on my second cup of coffee but it's not going down fast enough is the sun over the mountains yet it is yes it's just the rays are coming through the window so um kids are off to school i'm ready so i'm looking forward to this too this is going to be the first time i do a live stream with uh with a partner and uh i can't tell you how excited i am i'm a little nervous so let's see how this goes it's all terrifying um we are incredibly intimidating and the the audience is they they all brought their pitchforks and uh rocks to throw at us i appreciate that you know it's it's only going live to the whole world it's fine yeah so um so i have been as some folks inside of red hat know uh so one of my ancillary duties is to work with partners and netapp happens to be one of the partners that i work with for a multitude of reasons not the least of which is i used to be a netapp employee and i want to say that i am really excited to be here and i know that we've all used that word a few times now because netapp has recently announced something that is incredibly interesting to me it's something that i remember talking about when i worked at netapp three plus years ago so seeing all of this come to fruition is it's just it's it it makes me happy um so i adam i don't want to delay any longer i want to uh to let you talk about uh what netapp and red hat are doing together specifically around this little thing that we call astra um sure yeah let's just jump right into it so um let me start sharing my screen because i have got a couple slides that kind of talk through what it is that we're doing um and i will bring up the right slide here okay um so what have we built we have built a a kubernetes workload management system uh and what that means uh especially in regards to to netapp is that we are able to handle multiple different kubernetes clusters and workloads and our target our first target platform is open shift so that's where we support um this this application running we're going to open it up beyond that but coming out of the gate we're partnering with and and supporting only openshift for this first release so uh it's been a great partnership so far openshift is a fantastic platform and uh we're excited to to show what astra is and what astr can do so uh i i brought this slide to kind of talk through what the ecosystem of astra is uh and we're gonna focus on the area where you can see the red hat logos this is what i'll be demoing and playing with today but i want to talk about the entire astra ecosystem just so you get an idea of where netapp's going with this thing okay so this is my drawing this is not marketing material uh this is just what i put together to help convey this uh for as many people as as i can we we have pardon me we have uh astra two different kind of flavors of master we have the control service and the control center the control service is hosted in the cloud and anyone can sign up through astro.netapp.com to get an account and login to astra posted in the cloud and what it does is connect to google cloud provider or or azure and using your credentials so you give us your google credentials or your azure credentials and then we'll go query and find your kubernetes clusters and then what workloads are on those clusters we're going to call those apps and uh from there on you can back up you can restore you can clone you can hook up different storage back-ends and for the most part just watch your apps watch what your workloads are doing so where this becomes powerful is the ability to move those workloads from one cluster to another so maybe you need to put it in a different um environment a different um a different geographical place you need to uh just create a backup for um you know for safety sake and because you might be rolling out an upgrade and you want to uh have a way to get back so what asp does is it it manages your application and data at the exact same point in time takes a snapshot or a backup of that and then um can move that around based on what you're hooked up to so that's the control service the control center is the exact same thing but it's in your location it's in a custom on a customer's prem and like i said in the beginning we're only supporting openshift right now we definitely have plans to open that up and and we have verified it works on some other platforms but our support right now is for openshift uh versions 4.6 and above so adam i want to pause you there and i want to take a kind of a step back and a step up of so at the at the portfolio level at the the netapp level what what problem are we trying to solve and i think two where does or how does netapp and and open shift fit together and i'll i'll answer that second one first uh so and i'm working off of the assumption here that we have some maybe openshift folks who are new to netapp that are watching the stream or the inverse right netapp folks that are watching the stream for the first time and not familiar with openshift so openshift is red hat's kubernetes distribution right our goal is to provide a single platform openshift that can be deployed across many different locations so whether that's on-prem to you know physical servers bare metal servers to vsphere to red hat virtualization openstack whether it's to hyperscalers uh you know aws azure google and a myriad of other places um so we we our goal with openshift is to have a common kubernetes platform with a common set of services across all of those and then have that value add of you know not just kubernetes but all the things that make openshift openshift so kind of end-to-end platform life life cycle management for things like installation uh updates all of the administrator and developer tools right the administrator console and developer console and code ready workspaces and you know all of that other stuff um chris i think you posted in the link there to go over to yeah yeah so to learn more about that there for you yeah and adam please correct me if i'm wrong tell me if i need to stick my foot in my mouth so from the the netapp side what we're talking about is providing as you can imagine if you're familiar with netapp right persistent storage into those containerized workloads running inside of openshift absolutely so that's that's only one component though because really that's not what astrid per se is doing that's what trident does right and that's what that's what netapp has been really good at for a long time and trident has brought into the kubernetes world what about three three to five years ago and has been maturing so this is yet another evolution of our foray into cloud workloads yeah yeah exactly so trident was originally introduced several years ago as a way of hey i have some netapp storage i want to be able to consume that i want i want to use that persistent storage with my containers in kubernetes so astrid now takes that beyond just consumption right it's no longer hey create me a pvc that's x gigabytes of size with y storage class which happens to be serviced by trident it's now going into kind of the data management application management room that's exactly right yeah so not not only can you uh enable your netapp storage for your kubernetes workloads or your apps um that's that was already available but now we are going to manage both your app and your data um and and help you protect your system yeah and it's something you know for people who watch the stream you know that i have a regular uh stream live stream ask an openshift admin we regularly this past week or well this week we didn't have a stream because events will fast but last week last week we literally got asked you know how do i do backup how do i protect you know the data inside of my cluster this is something that comes up all the time yeah so i i hope it is interesting to folks to see you know how netapp addresses that when you're using the netapp portfolio with openshift yeah that's what that's what we're here to do that's what that's exactly what astr is is to gather your openshift clusters together gather all the apps that are on them and give you an opportunity to manage them and then with the click of a button you can snapshot them or back up them and move them around so this that is our answer to exactly that question i'm looking forward to the moving around part that sounds exciting yeah and here so let's see how the demo goes what else i appreciate you you backing us up a little andrew i kind of go right in and uh i want to make sure that we've got the context fully set what else should we address before we we dive a little deeper um i i think it's important to bring up that you know because for both companies you know hybrid cloud and you know the the multi-cloud and hyper scalers and all of that stuff is really important and you know with trident you're able to connect um what's the new one that that ffx rfrfsx fsx fsx so with trident and the various netapp services you know across those cloud providers even if you're using um and and i hope i'm not speaking out of turn here even if you're using like managed openshift you can still connect into those netapp storage services um if you're deploying you know user managed or or what do we call it self-managed now instances of openshift into the various clouds you can connect into those netapp storage netapp data services across all the various clouds and you can take advantage of things like well astra um so yeah i do i always like to remind people that um you know especially we see more and more survey results coming out more and more you know as much as we all love to read cio magazine and all of those other things you know about hybrid cloud multi-cloud um it is a thing and it works across both portfolios both portfolios work together for that so yeah exactly yeah we have a we have a great relationship here and we're we're going to continue expanding that so we have lots of plans for how how far we can take astra except when we accidentally kick you out of slack you know all right i'll quit distracting you okay so um i'm gonna i'm gonna drive back down into uh where i was going before and uh i have i have a really rough script of what i what i want to do here but uh for the most part as as was set up um as as our expectations were said as we going into this thing we're just going to kind of roll with it and see see how it goes uh so first what i want to do is show um the the diagram that shows you know astra is an application astro control center is an application that sits on a red hat open shift cluster and can manage other clusters so it can manage uh the cluster that's sitting on and you can add others and then when you add clusters to be managed we discover the applications that are there you can also custom define an app we'll get into all that um and then those apps either you've deployed them using whatever open chef tools you want um or config or cubistar cube cuddle well however you've deployed your apps is however you want to that we that doesn't matter we define we find your apps um and as you deploy more or take them off we recognize that and then um those apps hopefully are using um ontap storage uh through trident so that the pvs there that are um generated are um so they're backed by trident volumes uh by ontap volumes uh they are tried in pvs and they are then gathered and recognized together at the exact same time as whatever kubernetes resources make up that app so that's that's what we're going to do so let me recap just to make sure i'm understanding so astra consul control center is effectively a control plane where we can connect one or more open shift clusters into that aster control center that's right and when you connect in those clusters it effectively does some sort of discovery of applications to include persistent volume claims as you would expect yes where an application can be defined as you know kind of any arbitrary thing that you would expect inside of uh a cluster and i'm i'm because i've i've seen at least some of this before i know that you can define an application through like a label right any object and and this is something that's that's uh interesting to me it's not just the pvcs and the pv behind it it's it includes things like the deployment definitions and secrets and config maps and all those other things kind of any kubernetes object that has the appropriate label or i think it will even work at the whole name space level right space level yeah so let's let's get into it let's just let's just show that that's probably easiest rather than trying to yeah try yeah let's let's i mean we can talk through and point at those things here with the with the mouse but let's get into it so i do have one more slide i want to show but i want to kick off the installation first because it takes you know a few minutes so i'm going to start that now um make sure i have everything i need in place can you see the the words on the screen are they you might want to bump up the font yeah yeah it's well i can see it in the zoom here on on the stream it's a little small all right let me that's that's probably how's that there you go all right so let me clear my screen off and we'll list the the contents of this directory again okay so um this is boring this is just a uh a few yammels that are uh the describe uh resources custom resources so let me let me just bring up a text editor and we'll take a look at those things [Music] there we go okay so uh this is our resource document for an astra control center uh it's pretty simple uh it's got you know custom fields for various things what are we gonna call it what version number are we gonna deploy by the way i'm gonna for this live stream i'm deploying off of not the very tip of but our integration branch so this is actually stuff that's even beyond what we released uh back in august so um kind of a little preview even though it's it's it's not all the way out yet it's uh good enough for the demo i think uh we're gonna tell it that we want it to be a specific url so i have this um set up in my host file my internal um uh dns routing so that i can route this url directly to the ip address that it's gonna install on and uh some other some other junk here where are the images so when you when you get astra you get a tarball of images and you need to take those images and load them into your registry that's because we know that customers who are installing astro control center are putting them or bringing them on prem and usually have some kind of scanning requirements some kind of security requirements for images that are about to be launched on-prem we hope to eventually be able to host these images so that you can just click a button and pull them down we have the capability now but um product direction is every customer is going to want to bring them on prem and install them so i've already done that for this demo but we have to point to that registry so that we can uh launch it into our internal area now are we are we working on an operator for all of this we have an operator so that's the next thing we're going to show um that's i know that's first well i see that there's a crd right so i know there's a controller at least but i don't know if there's you know most of us in the openshift side we think of operators we think of operator hub um so being able to go into that you know very nice catalog interface and click and say you know deploy me in astra yes and actually we're making quite a bit of progress in getting our operator onto your hub so well we don't have it there yet um we have probably people watching right now who are working on getting that operator into your hub so that they can uh deploy it right into their open shift their their cluster and then do the things that are necessary to launch the acc using that operator so the good news is this process that will show you might not have to actually do here in a in the not too distant future you'll be able to just go in with to operator hub and and deploy it exactly yes so uh we we have a crd for um this just this describes astrid control center and this is a big long document of this is what the the operator puts in so instead of going through all of that let's just launch it so i have uh these two these i've called them apply because i want to apply these files i'm going to apply uh the operator first let me make sure i'm on the right cluster here um [Music] okay so i'm i have two clusters uh open shift z 0 1 0 and 0 1 4. i couldn't i have done i should have done sexier naming but that's that's what i have and we're going to launch astr control center into 0-1-0 so let's first launch the uh the the operator which is something hopefully you would do using operator hub um and it's as easy as apply apply this file which is what operator hub would eventually do i still get some warning about you know throttling from the from the uh the kubernetes api anyway so that's done so what did that do let's go over take a look uh this i have these two tabs here this one is 0-1-0 and this second one is 0-1-4 so uh 0-1-0 just deployed this namespace netapp eight it's too small isn't it i'm gonna say i'm gonna pause you for a second to yeah a little bigger it's a little bigger now you can click the hamburger up in the upper left to get rid of that side menu some of that stuff yeah all right so it deployed this operator and there is you know nothing really in it other than uh one pod it's got a pod in it and that pod has uh this this um [Music] container which is just listening for should i deploy an app on an astra or not and we haven't applied the crd yet so it's not doing anything yet so that's that um switch back here uh let's clear the screen list out my stuff again and then the next thing i'll do is now i can apply the uh actual i call it min because it's a minified version of the acc crd and then i'm going to tell it which namespace i want it to in oh i have to create the namespace uh oc create namespace uh netapp acc acc so um first thing to do get that name space created get the security context created for that and then let's apply uh this the crd and make sure that it's going to go into the netapp bcc namespace so ready go and all of this is if you've used you know crds before this is and operators this is pretty standard right of the first thing that we deployed was the operator which is really the controller behind the custom resource definition and for the lack of a better term the automation that implements that crd instance so now what you just did was say hey go create a new instance of the application which triggers the operator that controller automation to then go and deploy whatever objects make up an asterisk control center instance yeah go do stuff okay so um that's what it's doing the operator uh if i were to watch the logs here i have two two controllers the manager is the one doing all the work here uh is doing lots of things uh this log stream is just gonna flow because it's installing everything a better way to look at it is to switch to the netapp acc namespace and watch pods um we're not going to watch this the entire time because i do want to show um the microservices architecture diagram but what is happening is the operator is launching these pods and so it starts with a helm um a helm pod because all of our charts are our helm charts everything is going to launch all these micro services are launched using helm so we first we get all of our helm charts just to be servable into the name space and then we start launching the the framework the the bottom layer of the framework so influx loki nats console let's take a look at the microservices architecture so you can see you know the the pictures of these things that are launching right now it's funny to me how much the marvel universe seems to be affecting things like we've got thanos for the right metrics aggregator we've got loki we've got you know what i'm giving give a little preview of my uh search displays here of my demo but what if one of the things i was going to do is put images of marvel characters into a blog and then destroy them uh so well i can't get that any bigger i'm sharing my oh man maybe i could do it here there we go i can sort of do it here okay um i will zoom in on this uh as i talk about things there are two large boxes here they outlined in this dark gray and the largest one is acc i'll drive into it and the smaller one is a managed cluster so what services are we going to deploy into the cluster that hosts acc or the manage cluster that we just we we chose to manage so this is a list of both of those environments put together and let's just start over here on the left hand side of acc so uh i'm calling this the 21q2 astra control services diagram because this is what we released in in august uh in our q2 so over here on the on the left hand side we launched uh persistent slayer uh vault mongodb influx loki those are where we store you know those are the things that are stateful and then we have platform uh a core platform of services so the ability to uh have features turned on and off at future flags and settings um credential activity so credential not activity credential service which is interacting with vault which is where we store and encrypt usernames and passwords and cube configs and things like that we have an activity microservice which is just keeping track of things that are happening in acc uh we have some metrics uh this gets kind of complicated but for the most part all of our monitored sorry managed clusters and apps are sending metrics and logs into our system and we're gathering that here and keeping it for seven days we are also sending it to our hosted service cloud insights that's uh down here and let me make sure this makes a little bit more sense we have cloud insights um asap and cloud central which are involved in our ecosystem so we're constantly sending support bundles metrics and logs from your instance of asset control center up into our hosted services and we keep those for at least 30 days maybe more depending on how what kind of level of service you've paid for we keep seven days of them on prem but if you're hooked up to and can send things out to the internet you'll have a lot more um telemetry so this is those services that are handling all of that stuff uh for our uh user management our identity management we have um local accounts as of right now so we don't we don't integrate with anyone's oauth or sso yet but you can have local accounts in astro control center and use those usernames and passwords to log into it um we did we disconnected that from cloud central because uh we need to be able to support air-gapped sites so our first release was local authentication i can see you have a question here no i i i was thinking back to the auto support stuff of um all of that data all makes a lot of sense and it really helps the cloud insight side of things um cloud insights is not something i'm terribly familiar with i mostly see it during keynotes and stuff like that but the amount of information you can get and intelligence you can get out of that is really impressive um but no the the authentication stuff it makes sense um you know supporting disconnected installs which is you know of course we have quite a few you know disconnected openshift installs as you would expect um so no all of that makes sense it doesn't yeah i have a question about it we definitely are growing into more capabilities uh but right out of the gate we needed to make sure that we support um air gap sites and local installation people who are concerned with just this local um situation so our authentication and account management while growing is still simple and you have your own accounts you log in um there's some support services that handle gathering the bundles and sending them up to asap there's some licensing services that handle uh making sure your license is valid and what features are enabled because of that uh then we get into the real you know the meat of what astr does and that's app data management so we have um [Music] nautilus is basically our core service that's the one that is once you bring a cert a cluster on to be managed nautilus is the thing that hooks up the watches it talks to the kubernetes api of those um clusters and that's where all the discovery happens uh the reason it's it's surrounded by uh what's called composite compute or cloud extension is because we are going to a variety of different clouds we're either going to gcp or aks um we're going to red hat we're going to open shift we're going to uh local installations um we've tested out rancher so we have various different clouds or kubernetes worlds that we need to talk to and so we have a abstraction layer for that so the nautilus just does its job but it can talk to any flavor of kubernetes uh we then have the same thing for volumes so we go and find your volume information and and uh and then not also uses those to to gather data and application stuff and then it maintains it inside nautilus and then we have buckets i'll explain buckets here in a minute and storage provider or storage back-ends so who's the back-end where what data service is running the volumes what what not data what what storage service is running the volumes that are behind everything so we'll see all that work everything else is is uh involved with talking to trident or the the operator itself uh how does the operator work to to install stuff on a managed cluster there's some pretty simple things so telegraph and fluid bit are the main pieces of code that we install and they send data like i said to cloud insights or to the host or and or to the host uh in order to get those on and running we haven't we have an operator that turns them on and it configures them to send data to the right places and and use the right credentials and then we also have trident so trident has to be present on any managed cluster this is a reconciler essentially for trident so it makes sure that trident is at its latest version and is configured to talk to these the storage providers that we support and right now it's just on tap we're going to be coming out with our astra data store support next year too all right so i think that i've killed plenty of time and our installation should be done it's just waiting on the internet right to pull down all those all those container images uh it has them all um no it had it had them all so what this is there's a long list of of services that are running and you saw you know the diagram that represents them so this is this is all of them i should scroll a little slower uh what i look for is is traffic traffic is the last thing that we install it's the the you know the ingress so after everything is up and running then we open up the ingress and we're ready to go so let's go back to our console and see what's going on so if i uh get acc that is the short code for um master control center yeah use let's use the right name space so there is one cool that's that's nice let's see uh in order to learn uh if it's deployed yet uh let's pull out the yaml for it and we can look in the status and see that it is currently deployed so and the post install that which is after we deploy it we run a couple post install commands that handle certificate management because we're going to be behind a self-signed cert until you give us a search let's let's make sure that's all done and ready so we're done we're installed and the uh remember the the url oops that's not the one this is the url url i installed it on is redhat.demo astro.redhat.demo.com let me just try let me make sure one more thing is up and running before uh i attempt to visit that url so i want to make sure that the the traffic service um you know is that url is that managed through like an open shift route or is that some other config there no so that um we do have instruction for how to change that to an open shift route uh but because we are going into um [Music] other different flavors of kubernetes too i i know we're supporting openshift but we are um putting our own internal load balancer on or we're requiring a load balancer to be present on the host cluster we're gonna use that and direct traffic after it comes into the cluster so we've brought all of that ingress lower than a openshift route and then we have a knowledge-based article that allows you to disconnect that and bring it back up to an open shift route uh for for this demo at least while we have time i'm going to keep it the way it was if you'd like to show it transitioning we could do that no that makes sense um i i i think the um what i'm interpreting is that essentially it creates a service with a node port and then that load balancer just points to the node port yeah so it creates a service uh yes exactly so that we we've exposed an external ip this external ip is routed in on my machine i've i've routed the url dns router that url over to this ip address internal dns routing would be something that a an admin would do um or you would have uh the openshift routing in play here so let me make sure that my my hosts are aiming at so this is this is the traffic load balancer service and it's currently picked up this ip address and i actually don't think that's the right one so let me take a look at my post file oh oops still early so i wanted it to be on 27 7 and it showed up on 28. so this is the beginning of the one bit [Laughter] all right uh let's just go take a look at our in so what we what we use is a metal lb uh load balancer here let's go take a look at that system uh i'll get my hamburger back so i can go see the big map 34 and i wanted it would i say i wanted to go 27 yeah i wonder what would happen if i tried to just point it at 28. yeah i'm curious i actually i think 28 is actually used by a different machine on our network that's a good reason to be concerned [Laughter] the worse that happens i'm doing this demo and i just crashed a whole lab i'm sorry [Laughter] all right i've done that before i kind of don't know that yeah chris do you remember when i accidentally powered off my whole lab yeah yes that was brilliant oh it's serving oh let's see wow okay what is it surfing though all right oh stupid wow holy cow i hope there's not some machine that's gonna that part of the lab is working at least now yeah okay uh so the part that uh is important here is i deployed this remember this is our crd here i deployed it with an email of admin.astro.netapp.io so that's the login so let me make this bigger thank you sorry and the password the first password is the uid uh although you have to pre-pinned it with i don't think i can show it um let me do it in the text editor here so 8 capital acc dash and then the uid so that's the first password that you get and this is in the documents so i'll copy that text and put it in here and hit log in and the first thing it does is say all right change change your password so cool i'm going to change it to something that i am you know muscle memory can type here oops i'm sorry i'm sure it's not a password it's password123 [Laughter] all right we're logging in wow sweet it's up and running zoom in a little bit please yep if you don't mind yeah no problem uh is that enough or yeah that looks good yeah i couldn't read the the manage kubernetes add your applications okay welcome to live stream um admin astra so how did it know that it got that off of the crd too so if i were to switch back and you know here i told it its name the account name is livestream and my first name is admin and my last name is astra so it took all of that stuff off the crd and now it's it's deployed uh oh a big red banner oh no hey so what happened is it's been running for a little bit and some of the first things it does is checks on some stuff and there's no license so it's not gonna let me do anything uh you can see that the add button to add a cluster is unavailable because there's no license so i'll head right into my account and my license tab and but well so you it also hit put a notification so this is that this is the activity um service saying hey not only am i going to is the ui going to play display display a big red banner but i'm also going to you know make sure there's a persistent notification here that says whatever it is it needs to say so right now it's just no license phone okay got it all right i do have a license ready for this demo right here and i can apply it and so now i'm up i can i can host up to 4 800 cpu units that means i can manage clusters and however many i bring in up until i've hit 4 800 cpu units so that's how we do our licensing okay all right so back to the dashboard and now you can see well here's some note about my license yes i'm using an eval license and i could upgrade it after november 18th i i'm just surprised to see a storage product that isn't licensed by you know gigabytes capacity i wonder if there's any product managers watching this why they chose cpu [Laughter] uh okay so we need to back up our application so you need at least one object store buckets so this is where i said i would talk about buckets um what we do is we take your data and we take the the resource documents that represent your app and we we compress it all down into um a self-contained tar and then we put that in a bucket and we can put it we can do it in any bucket so um i have a bucket ready to go here but when you add buckets uh i did that kind of fast this is a list of of all the buckets that are available um when you add a bucket you have a choice for which one do you want a netapp on tap s3 obviously we're selling our own products here or a storage grid s3 or a generic s3 and so every s3 works um i've prepped an ontap s3 for this so i'm going to switch over to my ontap system i'm going to create a bucket and then i'm going to use that bucket again and so this is something that that i appreciate um you know i i've i've been a storage administrator of various types for the last 20 years and i constantly remind people snapshots are not backups so having this very distinct like yes you have to create a bucket and when you do a backup it's going to take that data and it's going to put it into this bucket like it's it's separating the two things apart snapshots are a point in time recovery a backup is a separate copy of that data that can be restored if the primary fully goes away absolutely that's exactly it so we do support snapshot snapshots but you can only recover on the same cluster but if you want to move an app to a different cluster we're going to need a backup because we're going to have to go to the bucket and pull it out um and move it over to that next place so let's add a new bucket i'm going to call this uh red hat netapp one and for some reason it has to be at least 95 gigabytes so let's put that in there so that will be done very soon and in the meantime let's go back here at the bucket name because i copied and pasted that put that right there so i have to go back to my other um this is tiny too uh so i have a confluence page that has what i need for the rest of the bucket adding stuff so usually you have some information about i mean if you know your ontap s3 system you're going to know what your endpoint is and what your access id and secret keys are um yeah so i keep that standard s3 stuff for yeah yeah i mean we have it for aws too like if you here's our aws bucket if you wanted to use that one should probably not like advertise secrets to our bucket it's actually on aws but there we go let's uh go back to here so the ip address the access key oops and then the secret access key and that's generally all i need so this will you have to put a name in here this is my bucket whatever it is a dead app on tap s3 this is the bucket name all right all good i'm just going to add that and it go it goes into a pinning state for just a moment and it's just going to verify there's a good bucket there and it can use that all right so that was um the dashboard so it as you saw it gave me a few prompts of things to do in order to to set myself up right so my prompts are now done i've got a license i've got a bucket let's get started adding a cluster so that's the first thing to do uh provide us with a a cube config essentially because we're we're not talking this is ask for control center so we all the only thing we need is we're on prem we just need to config to your cluster so we can manage you if we were in astro control service and we were talking to a cloud provider we would ask for your credentials for goog for gcp or azure but this is a demo of acc so give me a cube config so i have one for i'm going to bring in 0.10 that's the one that i installed it on i installed uh acc on so let's manage that cluster and then let's also manage one other cluster so the first thing it does is it says let's pick what your default storage provider is going to be so what it's doing right now is it's querying using that cube config to query the the cluster it's the kubernetes cluster and it's loading the storage classes so i only have one on there it's an ontap gold using the trident storage provider and it's eligible so that's good enough so it says confirm all of those details and i i assume in the docs it shows how to like how to generate a service account or a custom user to use for connecting all of that and generating the code config and everything it does so i'm gonna add one more let's let's click that link here in a second so this is gonna it's it's in a pending state right now because nautilus is querying that cluster and sucking down all the information about it so we'll let that ride while we does not deploy anything into the cluster it does it deploys a monitoring um namespace and then it deploys those uh telegraph and fluid bit agents i was talking about so that we can send logs and metrics back home in this case it's setting it to itself because it's i'm managing the same cluster i'm on but now i'm about to add yet another cluster so what what what would happen if trident wasn't installed to that cluster uh it wouldn't it wouldn't let you go any further it would say here are the storage classes you have but nothing is eligible to be managed by astra so it would give you instructions like it's doing here well so first you have adding clusters which opens our docs for adding a cluster um and talks about prereqs uh that's really small isn't it you should ensure that prereqs are met before you add a cluster and then it talks through a bunch of stuff you know what what export policies for ontap and check the trident version so i won't go through all that but yes we open up the document and say here's what you need and you were asking about the service account so that's these instructions here um where's my service account stuff here create an admin role cube config and so then it gives you instructions on how to do that all right i already have that so i'm just going to put the docu the the cube config that represents one four and we're gonna add that one a little moving a little faster now see i just um now i just need to figure out how to put it in rfe like can i can i add a cluster to be managed and have it deploy trident and then tell it to automatically configure trident for the storage system set it anyways automate the world man glue it all together with one command uh so i have these two clusters uh last thing let's we checked my my bucket was still pendings okay so bucket is now available um the clusters are coming in so look at this i've got some new notifications a storage back-end was added so when i added the cluster the 0-1-0 cluster it determines there is a there's a trident storage class and it's connected to a storage back end so we went into trident and we grabbed information about that back in and now i'm going to display the discovered back ends so i have this one here this is the ontap one um i told you that we're getting ready to do ads or ask for data store two so that cluster has ads added to it uh but i'm not gonna do that here in the demo because it's it's still um early in its uh development but because i'm not doing ga bits we're doing um development branch sorry development bits uh i do have it shown here but let's manage the uh the ontab cluster so i you have to put in your credentials for that ontap cluster and then import it and it moved then so oh it's successfully managed go to backends to manage it i'm in back ends but i'm on the discovered tab so now there's only one in discovered state if i switch to managed i've got this guy here who is my ontap ontap97 backend and it's available so i could actually go look at it uh it's still discovering more information it'll show me the capacity and here's all the pvcs that are on it and what um what cluster they are hooked up to there's a lot of stuff going on in here so so what i'm noticing what i'm noticing is you haven't had to do anything extra to connect astra to the storage system basically it's using trident to get all of that information so the the administrator's already configured trident astra is just taking advantage of that that's exactly right yeah we don't need you tell us your credentials because we're not going to pull that out at trident um but once we have your credentials we'll go ahead and watch that for you so you can take a look at what's going on i'll come back and see this if we have time so you can see some of the graphs that get drawn about it but um for now let's switch to the applications view so uh one more time the clusters that i'm managing are one o and o and four they are both um one dot version kubernetes version 1.220 um open shift version i think that's 4.7 yep okay so then i switched to the applications view and i have a discovered tab and it is discover it's discovered 46 things so um it's not displaying all of that um here's what here's what i want to do i have a wordpress website this is for the demo i pre-deployed a just the very the very vanilla wordpress website that i could using helm onto my 014 cluster and it is not managed yet so let's manage it so that we can we can get into it and start considering it a workload i care about these other workloads are uh well here's the operator um right now we're not managing ourselves we're and we're not managing any open shift name spaces sorry go ahead by default these applications are namespaces like it says i see one one to six entries but there's oh discovered is 45 okay i was thinking it'll just grab all of the name spaces and turn those into apps but um which is sort of true yeah we're also hiding a lot of stuff a lot of system level name spaces so let's just what's in here so if i open up netapp acc all of these i told you they're helm repo they're helm um apps so they're in the netapp acc namespace but they're a different helm label and so it gives you an opportunity to manage individual things like that so the top level is namespace you can add a label inside a namespace or outside a namespace and that would be considered an app so it tries to predict some things for you um but you could also define your own you could say um which cluster let's let's pick this one let's pick any name space i don't think i have anything in default i do have some stuff in default i don't know what's in there and then or you could you know select a label so or type a label so oh i don't want to manage anything in cube system please don't [Laughter] we discourage that on this channel yeah so uh here i i picked the netapp acc namespace and i could i could type out a label and it would narrow this down to um deployments that have labels that match that um so right now everything is selected and or is not sorry these are selected these are unselected anyway i can start defining custom namespace sorry custom applications based on a label and that's important it's a single label at this point um we are going to uh refine that a little further so that you could make an app out of just about anything that crosses namespace that's coming up soon too but for now uh this is how to define a custom app and this would open the health docs if you wanted to read more about that i'm not going to do it but just wanted to show it's available cool all right so i'm managing wordpress so let's just narrow my list down to what is managed i've got a wordpress site um this wordpress is posted on 014 so let's switch over to one four make it bigger shut down the hamburger and i'm scrolling to the bottom there is my wordpress um namespace um what service my wordpress service ah this is what i'm looking for this ip address is where wordpress is being served so let's just open up a new browser and put that ip address in and some plain old wordpress website uh it's been a while since i installed wordpress glad to see it looks exactly the same it does so let's go to um whatever that hello world post is let's edit it yeah wordpress got so fancy like i don't care but let's add something what do we let's insert something after um you can always paste those aws credentials back in yeah let's put the aws s3 pocket credentials in there no let's do that's dangerous let's let's uh iron man iron man images there we go copy that image and go back oops my zoom thing is in the way oh yeah little share screen box there we go update yeah put a picture of iron man in our blog post is updated so let's go back and take a look at it oh i have to open so i have to go into the post there it is there man that's really huge all right all right there's my here's my thing okay so let's go back to astra and uh let's let's let's back this guy up so if i go into first let's inspect what do i know about this thing it's healthy it's unprotected so no backups or scheduled policy yet it's running on 014 uh these are the images that it found about it um [Music] resources so this is where things get kind of cool it lo it knows every kubernetes resource that makes up that app so see these are cluster rolls some of this is open shift generated a whole ton of cluster oh man it's 110 things uh cluster roll bindings thank you openshift uh here's some secrets and i eventually i'll get to the freaking there's a stateful set there's the maria db or my sequel um there's some services that were created um there's got to be the wordpress service worker service somewhere there's a pod there we go here's some pods so anyway asterisk knows all of this stuff about your um application and when we take a backup it's going to go export the resource documents for every single one of these things and it's going to go pull all the data out of the volumes there are two volumes one's an eight gig and one's a ten gig for mariadb and for wordpress wordpress is store in the um the files that make up wordpress and mariadb is obviously my sequel so it's gonna get all that junk it's gonna press it down it's gonna put it in a bucket so let's do that back up yourself so well that's it which bucket do i want to use i only have one bucket it's the default bucket it's that one i created earlier and i could name it something if i want but let's just go ahead and call it whatever netapp or sorry whatever astro wants to call it um i think it's important to to point out just real quick that because it automatically selects you know all of those objects if you're using a route to access this application that includes the certificate that it's using yeah so in amongst all of those secrets and other objects there is things like the automatically generated certificates that certificate manager operator creates for those services and all of that other stuff so it's not just you know the deployment the pvc you know the config map that you created when you or helm created when you deployed the application it's all of the stuff that openshift also creates to go along with those applications which i think is really important right if um you know you restore back your uh your application in a different cluster and it has a completely different certificate that could break you know applications so yeah yeah there are things to be aware of look like like that um the service as you saw has is using an external load balancer ip of an ip that's assigned to this machine so when i i'm going to do that here in the demo when i move this thing to another cluster let's make sure that the ip of the service changed when it stood back up again um all right so moving on what happened there i feel like my data protection screen should have more on it maybe refresh the page that that tab isn't loading yes sub sub tab oh boy this data protection screen again i'm running off the the development branch that's strange it should list the the um backup that's being um created and yet last night it did so i don't know what it's doing here um this is the activity overview very strange um hold on a moment let's switch over let's switch back to the cluster so i can see if it's still doing a backup so if i switch to the projects close the hamburger menu scrolling to the very bottom so the bat when we do a backup we create a new namespace in that cluster and we hook up rustic which is what um creates all the backup files or helps us compress everything and put it into a different um put it into the bucket create all the stuff put it into the bucket uh that created a bucket called wordpress backup oh sorry no not a bucket a namespace so because that namespace is gone i know the backup is done but i am um that's what i was looking for here you could see the backup being created by no by seeing a new namespace show up for a short amount of time and then go away so it's done because we talked through that amount that amount of time but i'm concerned that i'm not seeing it well let's do something with this thing so if i wanted to move this to cluster 10 here when i say it says yes so here this is my backup it's done so i'm in a wizard to clone an app from an existing backup it shows me snapshots or backups i did create a backup the the ui for the inside of that app wasn't showing it but this wizard is showing it it says i have this backup already created which was done you know a few minutes ago moments ago so so essentially there's just some glitch happening with the other sections yeah yeah let's i mean let me see if i i can get it that's never happened before everything was going so well all right um i want to move it to the other cluster so i'm cloning this app before i go any further i could i could restore this app somewhere else i could the restoring clone are really the same thing um [Music] we use restore because the intent is you're gonna restore it right on top of itself we're on the same cluster and you can restore from a snapshot you can do you can dump it right on top of the same cluster uh but clone the the nomenclature the the semantics are we're gonna move it we're going to migrate it we're going to change its location so i'm even though you can do that with restore clones word uh means to me move it so we're still working out some shapes here so we're staking through the the ui here the ux like what really should we call this thing anyway i'm going to call it wordpress just without anything fancy and i'm going to clone it from an existing backup that i already took i'm going to put it on cluster010 because it currently only exists and here's some reminders on 014 and what its name is and a link to the document so let's review that and get it going all right so what happened well it automatically has a new managed cluster i'm sorry cluster a new managed app so wordpress showed up and is in a discovering state on the 010 cluster so it's there's things going on under the hood let's go see what they are on the o 1 0 cluster scrolling to the bottom here here there's a new wordpress um namespace on that cluster and there's nothing going on in it yet let's close the hamburger menu and you can start to see some activity so what it's doing is it's loading all of the volume information into the new volumes let's show that we provisioned volumes we're loading the data on we've attached to those volumes we're loading data onto those volumes and there's some jobs running the in the background these two pods are they're they are the r means restore they're restoring these things into the right volumes and then one once they're done they both will finish they'll go away and then the resources will be created um and by resources as the pods the secrets all that stuff it's going to show up so it takes a little bit of time here so just to be clear we're not we're not doing or we're creating a new volume and then we're loading the data into that volume yes okay that's right and then that's effectively what those are our dash or our hyphen pods are doing yeah so these are pods created the volume attached to it um loaded the data into it they're gonna disappear um and then the and then astra will create the other pods that are actually the wordpress mariadb and the wordpress itself so you see that happening right now um and out of curiosity does it does it back up like container images or are those we we expect those to be available externally uh yes good question no we expect those to be available externally okay back up the registry then there's a lot of stuff that could be in there so yeah we expect because what's going to happen you you installed this in your environment and you pointed to an image path and it was able to come up before the the exact same image path is in these pods and we're expecting the registry to have those images all right so things are all getting going here so mariadb is now up and running it's ready and now wordpress will come up and running let me bring the hamburger menu back up because i want to take a look at the whole sorry scrolling fast so it's doing stuff it started the container yep i see the config maps are back the secrets are back yep so is i was reminiscing because our our last stream was on api deprecations about way back when stateful sets were called pet sets yeah i didn't even know that really yeah back in the like 1.4 in earlier 1.5 something like that we've come a long way haven't we what does pet set mean because the a the thought process was a pod that has persistence is more akin to a pet than a you know chicken cattle bacteria whatever analogy you want to use so it was it was a heavy-handed andrew's opinion heavy-handed way of of saying that you know this one is is special staple sets are not special date is life all right so um while you were talking i was bringing up the wordpress service and we wanted to make sure that the ip address so this remember my last blog was on 37. um now because this wordpress service was a is a load balancer it looks for um some internal load balancer to give it an ip address and so when it came up on the other cluster the load balancer on that cluster gave it an ip address a new ip address whoops tab of 34. so serves the reason that if i tried to go to 34 240.34 my blog should come up uh-oh here we go with our fun so what's going on here i wonder if wordpress is all the way up yet oh good point because it does take a little bit of time yeah where are thy all right well so something's going on with my uh 34 web oh [Music] all right let's troubleshoot that but before we troubleshoot that um i'm looking at sorry i i'm looking at the old one this is 37 i shouldn't have shouldn't have switched to that screen still waiting for a 34 to come up i'm just going to shut that off so it's not distracting and i'll figure that out and here in a minute but what i do want to show is that this wordpress app that's on oh she's actually should have named that something so that i could so because the only the only difference here is this four and this zero yeah anyway you can see it's a little newer it's discovered a little bit more recently it's not protected because we've taken no backups of it it's on the cluster i moved it to so let's go take a look this data protection screen still isn't coming up so i'm a little bummed about that but let's go if you look at all the resources there are 122 entries of resources here um that it loaded it created for us um on the openshift cluster it moved it to the storage has two pvcs uh they are you know on tap gold eight gigs and ten gigs so it basically it cloned the whole thing it just moved everything over and now here it is um so adam is is there a way to do effectively you know manage this um application protection application you know backup process through an api oh yeah absolutely yeah okay cool uh you're distracting me a little bit but this is really fun uh so uh oh by the way i went to my account i'm using 48 cpu units because it knows those two clusters have um they each have 24 cpus so if you go to your profile in this upper right-hand corner there's an api access link if you click that it's got the this is the api documentation which is a swagger dock and that loads our entire api for you to play with um you can just click it and start trying it out right now but you're not authorized so in order to authorize the swagger to work you would generate a token uh it wants to call it tan pasta that's weird or broken rhubarb whatever um i'm gonna copy this and close it and it'll tell me i've got this api token it'll never show me that again so i copied it into my clipboard and then i can go here and click authorize and paste that in and now i can use my swagger doc to discover the api then obviously you can you can um yeah and what i'm curious about here is essentially as as an app admin as an app team you know can i avoid the gui entirely either for openshift or astra and what you're saying is yes absolutely you know you can deploy an application you can protect an application you can restore an application and you can do it all from you know all programmatically using whatever tools you're already using today yeah yeah absolutely that's that's the point uh so i'm going to call for all the apps what what apps does astrid know about in this in this account account is just the the tenant you only have one in acc and it here it says it's showing me everything that it thinks is an app remember i haven't custom defined anything um and there's ids if i wanted to go learn more about each one and see if there's anything managed uh going really fast probably should have this is there's a ton you saw how many things it thought was an app no i i i think we get get the premise i just wanted to check and make sure that or or check and see if it's possible you know to do all of that um again i think a lot of especially application teams um unless they have to don't want to use a gui so it's great to see that that that is a an option yeah absolutely and we have a we have a whole document on our api and in addition to the swagger docs we have um regular docs that are published to describe the whole thing what was i doing looking at apps this one's not protected why is um what's going on with my 34 um what could be going on are the uh are the first three octets right in that ip address 10 193 240 34. so yeah yeah um all right i i need to take a quick bio break do you guys mind i figured yeah we can take a break if you need to yeah no problem this is a short break i'm gonna put um i don't know you guys take a quick i'll be back real soon i'll put astr back up and uh yeah sorry right back i mean the routing is which it looks like it's set up um that should work hey let's yeah let's do a rodeo style let's figure out if that works open actually wordpress might object depending on the wordpress deployment like oh no you're coming from the wrong route i'm going to redirect you around wrong url but yeah well there's no url associated with it is there well there's got to be some kind of so this is the one this is the first one that's running on this yeah well so something inside of wordpress is like i would think is providing the url i don't know uh oh it thinks it's here it thinks it's 37. this is the this is the one we took a backup of so i don't oh i mean what's creating a route only takes a second so what's the worst that happens yeah let's let's let's see what happens here all right so create a route on 14 no 14 was the where we started we moved it to 10 and um we want to be in the wordpress project so no routes found all right you guys talked me through it so just hit the create wrap button and then uh call it whatever you want my route word grace yeah or no you you'll actually have to fill it in oh let's just call it wordpress yeah uh host name you can ignore okay path should be fine service will be the wordpress service select the port and hit create and then uh click the location link there boom hey hey amazing and then it should have yep so there's that picture of iron man that i put in before i took the backup there you go nice thank you andrew yeah it's all magic [Laughter] oh yeah okay so that's astra has cloned an app from one to another and i taking a look here i still don't know why my data projection screen isn't coming up but i can i could i could take a backup of it now i can move it again i can do a lot of stuff but let's kind of take a tour of other parts of astro see what else is available so remember i showed that we have these the back the storage back in this is the on tap that was hooked up because of trident and now i should have a little bit more information about it so i know what capacity it has i know what capacity each and every volume is utilizing so 1.65 of this maria db volume is being used that's kind of interesting information um this activity is asked just that we added it and managed it so far and there usually is one other graph that's not showing right now um that shows um utilization rates i'm not sure why that's not coming up but it's not other stuff that we have here um in the account screen you can create your users this is where you create users by the way so this is the one that was created when i deployed um it's you know just as simple as this screen right here first name last name email address temporary password because the user will have to change it the first time they log in and what role so we have four roles set up right now in astro the owner can is the owner of the entity let me pause you for a second adam oh no okay i was i i it was a mistake okay sorry yeah no no worries i was going to point out like wait my stream just disappeared no it's back now so there you go sorry they could still hear us okay so we have four roles um owner and admin are are similar um when it comes to astro control center you can create users and um and control their access a member is a user and a member uh is assigned to um an account in this case in azure control center we only have one account but that member can then um create backups and and restore apps and things like that i just got logged out huh that's interesting somebody found your ip [Music] uh astro.netapp.io and then i believe that's the password i set it to yeah all right where was i i was creating a user anyway uh or viewer a viewer is just someone who can you know navigate around in here but they can't create backups or restore apps or anything like that credentials is just a display of things we know about our world here so um we have a username and password to access the ontap back in remember i typed that in so it stored it we have the cubeconfigs for the 010 and the 014 clusters and you can do some stuff with those like some of them um you can't remove this credential unless you unmanage the cluster but this one you could remove i don't know why you'd want to it's a it's just a username and password that's used for the storage backend but we we intend to expand the capabilities here so anything that is secure it's the the data is actually stored in vault and it's encrypted but here we show you a list of those things so that you can see what is what is in play notifications is anything that has been going on here and those are also available in this little pop-up right here and mark them as red and mark them as red in this notifications area um this is the license we already took a look at this and here's here's a real interesting one so connections we're going to be adding more ui cards to this as as more functionality comes but here's what's available now a proxy so remember we said we're going to ship asap bundles and metrics and logs up to cloud insights well if you have a proxy that you have to go through to exit your your intranet you can connect that in right here you know for the server name the port and then if you need authentication to do that so that telemetry works this is cloud insights if you want to connect to cloud insights so so that we're sending the data there you would um go to that tenant url uh whenever you you have a cloud insights account you have your own personal url and you would put that in here and you would have gone to there previously and generated yourself an api token one that's capable of reading and if it's the same one is also capable of writing you could you would put that in here the same one in here or you could generate two and then you'll hit connect and then we'll be sending data up to cloud insights um [Music] fluent d if you'd like us to ship your logs to a fluid d instance in your internal area then we'll do that so we already have all of our logs coming in to our own area and we're going to be sending stuff to cloud insights and to asup but most kubernetes systems have a log management area so you're probably going to put it in elasticsearch or logstash or something like that so this is how we feed our logs over to that so if you're internally wanting to manage your stuff as opposed to sending it to cloud insights or both this is how you turn those things on yeah and that's similar to what we do with openshift right with the log forwarding api and all that other stuff so but these are going to be more astra concerning logs like right now i think we're doing events and notifications um [Music] so these are astral logs but then again if you astra is an app on an openshift cluster so you might be able to get some logs that way so you you have a variety of options for how you want to manage your kubernetes ecosystem yeah i appreciate that it was thought ahead of hey you've probably already got some sort of logging and alerting system that you can just hook right into um you know if you don't want to have that external system configured or uh external meaning cloud insights or or whatnot exactly yeah so we have our own system you might have your own system you might be behind a firewall or you know behind a proxy so we're a lot of we've thought of what you know the bare minimum we definitely have more ideas coming for integration points uh but this is this is the beginning well there's always more work to do right oh yes actually yeah yeah i've got plenty to do today too uh activity this is a tab of just or a list of just everything that you've done so far so if i if i back up when i deployed this it created a user for me it added that user um and credential then i logged in i created a license uh it's it noticed that it um [Music] when i added a so add credential i think that's when i added the cluster um it deployed the monitoring agent onto that cluster i didn't even show that i should have shown that so both of the openshift clusters now have another namespace called netapp monitoring and in there let's look at the pods are the fluent bit and telegraph pods that are shooting metrics across the board anyway i don't have to go through the whole thing but this basically is a script of the demo i just went through um things the system was doing as i was managing an app cloning an app um cloning an app uh i locked somehow it logged me out and i logged myself back in so anyway that's about what i had planned for showing so um i'd love to you know go further but we are about an hour and a half into this thing i do have more time if you guys want to see other things you know it was only a little bit of stuff that you showed there um so i i think you know and just to recap pretty impressive in an hour and a half time with a whole bunch of other introduction in there that we literally you know deployed astra protected an application backed up an application and then cloned it over into an entirely different cluster um you know the only thing not that i think that we you know need to show it um but the only thing that like rounds that out is like taking a snapshot um and it kind of encompasses all of that data protection aspect whether it's point in time recovery with snapshots whether it's actual backups you know to an external and you know s3 bucket you know all of those different things so pretty impressive within the amount of time i'll default to anybody who's on the stream if you want to ask any questions uh if you want to request anything that we can see or or that we can show after the stream ends you are more than welcome to reach out so i i tweeted about this stream you can in that tweet you can find uh adam's twitter handle which is at adam hade mine is at practical andrew you're also welcome to reach out to me i'm i'm pretty easy to find uh firstname.lastname redhat.com andrew.sullivanrad.com so yeah same for me uh firstname.lastname.netapp um my i retweeted this too so at adam hade had also retweeted the exact same thing and um available on slack through netapp and red hat channels usually um okay so while we're letting people kind of digest that and decide if they want to ask questions let's take a snapshot because we haven't done that yet it's the exact same process as backing up except i don't have to choose a bucket to put it in i just say take a snapshot so we're going to call this one whatever astro wants to name it it looks like there's some convention here with dates and then some i bet it's a date and time review and go so again i'm a little bummed that my data protection screen is not showing there's a and does that does that create a yeah yeah there isn't some incognito thing yeah i could so this is my fault because i wanted to show some of the latest bits that we've got here and i i pulled this off the development branch um in order to get some of the cooler things that are that are coming um particularly the backup screen when you when you go to take a backup you have um some new ways of displaying things and also the the storage back-end had some really cool things with the graphs that are being deployed um and capacity information but because of that there's a there's an error here in the ui that uh i'm not showing data protection um i'd have to i could in and deploy a whole new astra and hook up these exact clusters to be managed again uh and show um with the ga bits so i'm a little bummed about this but um i everything is still accessible to me through other menus okay so that that snapshot is that a csi snapshot as well like could we do uh like from from the openshift gui can you see those snapshots if we create one i assume it will also like if you create one through openshift would it register or would astra be aware of that no astro only knows about the snapshots it took okay um but and that i'm assuming that's because astra is snapshotting more than just the pvc it's probably snapshotting the state of the other kubernetes objects yes yeah okay yeah because that actually makes a lot of sense um you know if somebody goes in and and you know changes the password for the database on this wordpress site and you go to restore that back well that password is probably in an environment variable or in a secret hopefully um you know so you wouldn't capture that just doing a pvc snapshot whereas an application snapshot would have the state of all of those other objects as well exactly yeah so here we're displaying it in openshift these are the snaps this is the snapshot that was just taken so we've got um the snapshot of the uh pvcs and then net um astra is has additional information uh for point in time for all of the resources at the same the same moment uh that this application snapshot was taken but it wasn't packaged up and put into a bucket like a backup is yeah yeah that makes sense so this is um on 0-1-0 this was the restored version of the wordpress blog and then i took a snapshot of it the original is still over on 014 and there should be snapshots there because like i said a backup is a snapshot that's been then compressed and moved out to a bucket and so there should be snapshots on oh no it was moved we moved it into a backup and moved it out so yeah there we go oh that's interesting so when you do a backup it creates a snapshot backs up that snapshot and then deletes the snapshot that's that's that looks like that's what happened okay i can i can see and appreciate the decoupling of those things there right because the snapshot gives it that point in time nothing's going to change from here and decoupling from there and the backup is portable you know you want to be able to restore this thing somewhere else where you don't have a history of snapshots it's just this this is everything so we took a snapshot to turn it into a full-blown backup and then moved it into a bucket which is then movable to another location all right um any questions come in or we got something coming in here oh a lot of follows yeah just thank you yeah thank you for everybody following um we appreciate all the all the new netapp folks so um yeah a couple of uh thank yous and good jobs um but yeah thank you to our audience um adam is there anything else that you want to show anything that we can uh highlight you've been you've been extremely thorough so i feel like the answer there is probably going to be no no at this point well what i'm tempted to do is start talking about all the cool stuff that we got coming but i don't want to make promises or or give dates but as you can see this is the beginning of a hugely um growth capable product so you know not only are we going to be working with openshift we're going to work with all other flavors of kubernetes we're going to work across public and private clouds we have a way for um you know moving your app up to aws or back down and we've got you know so that bridge is coming into play we have um data so data protection so um data sorry dr disaster recovery uh it's all acronyms uh so we've got dr coming up so that if an app crashes astra can be watching other places where you might want it to stand up and manually at first can stand it up somewhere else because it's already been synchronizing things we'll be using netapp's core you know data data um protection capabilities to be moving apps moving data faster than um our current the current one you just saw so you know we've got a huge list of improvements to make to this system yeah i'm uh i'm curious about and i'll ask the question for christian about like git ops and whether or not there's a possibility for get ops integration around some of the the data protection aspects um although that would probably depend on having crds for many of those things yes well so almost everything that we're doing uh is in intended to end up as a cr crd so like we have we have the rest api that um can be activated by crs um and you saw a little bit of that at the beginning uh but for regarding git ops we've been evaluating it or thinking about it so we we we have the list of things we actually are going to tackle and then we have the things that we are considering and so that was that's probably out there in the considering we have had some conversations about it but haven't um haven't made any commitments yet makes sense um okay so i don't i don't have anything i'm i've been looking at our you know our shared doc here that we did a lot of the planning and i think all of the things that i had thought of beforehand we've addressed um in significant detail more than i was expecting so thank you yeah there's been a blast guys i really like doing this stuff um yeah it's funny you know we've been streaming now for what a year and a half chris and you know we my my stream just crossed a year old and it's it has been it's a very different environment but it's a lot of fun um and you know it's not like a conference you know we're i think we're all used to going to conferences and you present and everything is planned months in advance and you rehearse sessions you know some people do it dozens or hundreds of times and you know you expect everything to work exactly the way that it always has and live streams um it's uh it's interesting because things break all the time you know chris who's on you know ninety percent of of the live streams has seen many things go wrong um both in my life yeah [Laughter] environment matters too yeah and it's it's a lot of fun um to troubleshoot those things or to kind of um go gosh i'm gonna i'm gonna sound old go with the flow i guess i am old right now it worked i mean we did it today you we created a route because the ip address wasn't routed properly so i mean that it's exciting but it you know it didn't come crashing down too hard i'm kind of happy about that real real time troubleshooting always fun good stuff um so chris if you uh so next week my show will be back at its regular time we'll be talking about uh bare metal installs um with reese oxenham so it's always more official when it has a british accent associated with it so there's exactly yeah i'm working on that i'm not gonna you know no yeah that's true you only want to go with the australian but it's better for me um but yeah i so chris i don't know if there are any other streams you want to highlight um uh i mean the level up hours next week obviously same day as your show uh before it at 9 00 am um we're doing the last call for code for racial justice on tuesday next week uh devnation will have their normal thing on thursday we're doing a what's new briefing for openshift49 oh yeah thursday morning that was the one i was trying to think of yeah starting at 10 a.m eastern 1400 utc so it'll be all the pms will join you can ask us questions we'll get them answered for you it'll be a very interactive session and you get to see what's coming out the same time everyone else does should be awesome um so i'm just posting into the chat here a link over to the astra control center landing page over on netapp.com or cloud.netapp.com um we also did so i along with the product management team over at netapp we did a webinar i don't have the link to that um yeah i'll have to dig that up and we can share that but um so we did do a webinar where we talked about many of these things at a high level um you know andrew's andrew's technical as i hope most folks know so i've really enjoyed getting to see things in action and seeing the uh the low-level stuff here um but yeah if you have any questions don't feel or don't hesitate to reach out uh practical andrew on twitter or andrew.sullivan redhat.com um i will also throw chris under the bus at chris short on twitter or short redhat.com and adam i'll let you volunteer your your public contact information yeah absolutely anytime anybody wants to talk about this stuff i love too so adamhaid on twitter or adam.hate netapp.com and um i it's been a pleasure guys this has been a ton of fun and i would love to do more stuff like this so as it comes along i'll obviously let you know and maybe we can do another show and tell absolutely great all right well thank you very much to our viewers uh i hope you have a great rest of your day a great weekend um hope for those who don't have to work or continue working over the weekend if not stay safe out there everybody [Music] you

Channel: OpenShift

Views: 472

Rating: 5 out of 5

Keywords: Red Hat, Red Hat OpenShift, Kubernetes, NetApp, NetApp Astra, OpenShift, Astra

Id: FmSxI59nY_U

Length: 105min 5sec (6305 seconds)

Published: Fri Oct 01 2021