Season 2 — Episode 3

Have you considered putting your backups on auto-pilot? Rebecca Fitzhugh, Director of Developer Relations at Rubrik, discusses the modules that leverage Puppet on Pulling the Strings. Learn how to automate your Service Level Agreements and a simple way to install Backup Connectors across a fleet of servers simultaneously. After listening to this episode and grabbing the module off of GitHub, your boss might consider giving you an extra vacation day. Be sure to stay until the end to learn about a recent Ransomware attack and how immutability gives you control to recover quickly.



00:00:18 Demetrius Malbrough Hey, everyone, thanks for joining this episode of Pulling the Strings podcast powered by Puppet. And I'm delighted to be your host. My name is Demetrius Malbrough, principal technical product marketing manager here at Puppet. And I am really excited today to talk with Rebecca Fitzhugh, director of developer relations at Rubrik. How are you today, Rebecca?

00:00:42 Rebecca Fitzhugh Infinitely better now that I'm talking to you, Demetrius.

00:00:44 Demetrius Malbrough All right. Fantastic. And also, let me give you a brief introduction of who Rebecca is. And so, Rebecca, she is the director of developer relations at Rubrik. And prior to joining Rubrik, she was a consulting architect focused on private and public cloud topologies as well as DevOps. And today we are going to talk about one of our automation integrations with Rubrik. And Rubrik provides multi-cloud data operations and a powerful, policy-driven platform to simplify data protection and unlock insights from data residing in the data center and also cloud. Rebecca, you ready for the questions?

00:01:26 Rebecca Fitzhugh As ready as I'm ever gonna be.

00:01:27 Demetrius Malbrough All right, let's go. So I guess let's start off really simple here. So what are you working on these days at Rubrik, Rebecca?

00:01:35 Rebecca Fitzhugh I have to tell you, it's not interesting stuff at the moment.

00:01:40 Demetrius Malbrough Oh, OK.

00:01:40 Rebecca Fitzhugh Right now, I've really been focusing on our go-to-market strategy. But the good news, as my team's actually getting to have a lot of funds, they've really been digging into a lot of CI/CD tooling and being able to integrate our platform with the CI/CD to be able to provision and protect cloud workloads in one workflow, which we can do on-premises. But now that we're rolling out more and more cloud protection, we really want to make sure that we have that same experience there.

00:02:04 Demetrius Malbrough OK. So I know you love to travel. So any travel since since COVID-19 has hit, or maybe you've just been laying low and kind of staying put, which is not typically your style.

00:02:15 Rebecca Fitzhugh It's been a big change. Yeah, no travel. So definitely a lifestyle change. So, I actually realized the other day that this is, I think, the longest period of time that I've ever been home as an adult. It's a lot, but it's really kind of giving me a chance to try on a bit more of a kind of a pedestrian life for a few months. But at the end of the day, it's about really kind of prioritizing the health and safety of myself and others. And so I'm being a good kid and staying home.

00:02:39 Demetrius Malbrough Well, fantastic. So so you mentioned a little bit about your team. I love the team. It rocks. I actually work with the team as well. So I have a little bit of a Rubrik history behind me as well. So tell us a little bit about the team and what are some of the cool things that they are working on right now?

00:02:57 Rebecca Fitzhugh Yeah, so if we kind of take a step back, you did a quick introduction of Rubrik. So Rubrik is a cloud data management company. That means we can back up and recover your data, replicate your data archive for long term retention with automatic expiry based on that retention threshold, and we even provide additional data services such as data classification and governance and anomaly detection for things like malware, ransomware, or a rogue employee going and deleting a bunch of data, and so on. And we do this regardless of whether that data is on-premises or in the public cloud. And so at Rubrik, I head up our technical advocacy group, which consists of two teams, so our team of developer advocates and our team of solutions architects. And so the Dev team, which is I know I'm never gonna change my title unless I have to, I'm going to keep  claiming that team because I love them so much. Well, I shouldn't say that, I love the solutions architects too. Well, you know, the Dev team is really focused on our core platform and our APIs, whereas our solutions architecture team focuses on our huge ecosystem of data sources. And so when you think about what cool things do we get to work on every day? It changes. Every day's a bit different. And that's what makes the job so fun. But what we're really focused on at the end of the day is enabling our customers. So this means demoing new use cases, helping grow our community and helping build tools that leverage your APIs to do more, and to be a little bit more specific, we've really been kind of rolling up our sleeves on Office 365. We introduced some new protection there recently, as well as GCP and RDS, so AWS RDS. So that's what we've really been kind of working on and focused on the last few months.

00:04:34 Demetrius Malbrough OK. You guys have been really busy over there. So I'm actually looking forward to seeing some of those things roll out, especially the O-365 piece and also GCP as well. Let's talk a little bit about the Rubrik module for Puppet and what exactly does it do?

00:04:51 Rebecca Fitzhugh So our Puppet module enables customers to easily install a connector in an orchestrated workflow. And so the module's really kind of allowing that interaction with the connector configuration with respect to our SLA domains. And it also allows the connecter to self-correct based on the configured SLA domain in the hierarchy. And so the connector, what that's really allowing Rubrik to do is to be able to talk to more and more workloads. Even if there aren't necessarily APIs available, we can install that connector and manage the data there.

00:05:19 Demetrius Malbrough So installing a connector, so maybe think like a Windows and Linux server and being able to install them on that particular system so we can actually go and have that deeper integration as far as taking a snapshot or taking a backup of that particular system right?

00:05:34 Rebecca Fitzhugh Exactly.

00:05:35 Demetrius Malbrough Let's go a little deeper into SLA domain. So that's service level agreement. I don't know why anyone would not know what that is in IT, but let's just talk a little bit about the enforcement around SLAs and the SLA domain. Why is it important to have SLAs to protect your data anyway?

00:05:52 Rebecca Fitzhugh Yeah so, really, when we think about the genesis of Rubrik, it was really kind of around this idea that it's inefficient and extremely complex if you're managing your data, using these kind of manual imperative jobs. And so in order to really kind of start treating data management as an asset, we really believe that you first kind of have to dispose of this concept of backup jobs as a unit of work. And so the idea of using individual jobs for each kind of data management component is really a model, if we think about it, kind of akin to asking a team of people to route packages all over the world for, say, FedEx or UPS instead of just using computer algorithms. And so we really wanted to get away from that model. And so instead of managing workloads in kind of these job silos, we imagine that there is a better way to do this using declarative holly policy based protection. And we can take these policies and broadly assign them to ensure these different service levels, right, for that data protection and data management. So this is really tying back to the business logic that our enterprise customers already had in place, right, they already have SLAs, SLOs, RTOs, RPOs, so why don't we just take that and then define this in a declarative policy and then assign it at a top level of a hierarchy. And then just let the computer do the work for us, right, and rely on that set of intelligent algorithms. So, if we think about that FedEx scenario I gave a second ago, it changes that kind of idea to really being able to pick between, say, a priority or standard class service based on how fast you're going to need that package to arrive at its destination.

00:07:24 Demetrius Malbrough Well, you definitely getting a lot better and more crisp with that story that you're telling, because I've never heard the FedEx analogy, Rebecca.

00:07:31 Rebecca Fitzhugh I would like to say I came up with it, but I definitely heard someone deliver it. Not quite that way, I was like ooh, I think there's something there. And so I took that idea and I sat on it for a while and I finally kind of came up with that. I was like, all right. I think this is actually a great way to explain it. Like, why would a bunch of humans sort and do all of this work rather than just using a computer algorithm, right? When I think about going and sending a package, I just want to say get it there as fast as possible, or I don't care, you know, the cheapest possible service. And so with that, I think that really resonates because everybody can understand that. And data management is extremely complex inherently. So when we kind of use this analogy, it really makes it, I think, kind of crystal clear.

00:08:11 Demetrius Malbrough So we talked a little bit about the Rubrik connectors for, let's say, Linux and Windows kernels, right, so can you tell us why this integration is important and maybe some of the challenges that would alleviate.

00:08:24 Rebecca Fitzhugh Yeah. So when we ingest data, you know, from like cloud workloads or from the, we use native APIs to do so. Well, when you start thinking about things like application level awareness and consistency or, you know, being able to manage data and physical servers, there's not always a public API available to use. So therefore, we use that connector to allow Rubrik to interact directly with those operating systems. And once the connectors are installed, they're automatically managed and upgraded by Rubrik. It's kind of hands off from that point, on autopilot. But that installation process can honestly be a pain, especially for enterprises with thousands and thousands of workloads like this. So a very common use case that we see that customers do all the time, and especially with Puppet, is using it to really kind of create that automated deployment of those connectors.

00:09:11 Demetrius Malbrough OK. So, you know, how does utilizing Puppet along with, I guess, the module that you've created, how has that made it easier to  install the connectors?

00:09:25 Rebecca Fitzhugh Yeah. So what our customers can do is grab this module and then create an automated workflow that says, you know, here's where the connector's located in terms of the package itself. And I want to go ahead and push this out to a thousand or ten thousand Windows or Linux physical servers. And it really makes it kind of one click scenario where they say "go," after they've written a few lines of code, and then we go ahead and we have that provision installed, which is just a huge timesaver. And it ensures consistency, which is key across all of those workloads. You know, you didn't accidently miss one, which I guarantee, Demetrius, you saw the lab environment that we used to share. You know I'm gonna forget something if I have to do everything by hand.

00:10:04 Demetrius Malbrough Yeah. And so I'm gonna put you on the spot a little bit here, Rebecca, since, you know, we go way back. So are you planning on creating, like, more Rubrik modules for Puppet? And if so, do you have any thoughts around what you would consider automating?

00:10:18 Rebecca Fitzhugh Yeah, absolutely. I'd love to, you know, but I would kind of say the disclaimer here is that we really create these integrations based off of customer demand. So if you're a Rubrik and a Puppet customer and you're listening to this, let us know. I would love to see more Puppet modules for workflows, like backup validation frameworks where you're live mounting a snapshot from yesterday and running a series of tests and then just throwing it in the garbage when you're done and having that kind of comfort at night that, you know, your data is recoverable and restorable. Or, things like test automation, so, for example, we have customers that clone like a production database and then mount it in a test environment using Rubrik in order to use it for QA purposes. So I'd love to see more use cases like that being Puppetized. And so if you're, again, if you're a customer, and you're listening, tell us. Let's build it.

00:11:05 Demetrius Malbrough So it also sounds like it's an opportunity for us to work together again, Rebecca, maybe a joint white paper or blog post or something like that.

00:11:14 Rebecca Fitzhugh Maybe I shouldn't say this in case the rest of the team listens to it, but you are my favorite person to co-write with, so I'm always down, Demetrius.

00:11:22 Demetrius Malbrough And I can get it, you know, I can get really, really, really creative too.

00:11:26 Rebecca Fitzhugh I know, I was actually recently rereading a paper that you and I had written, I don't know, three years ago, it was about our kind of core actual piece of data management, a component called Cerebro. And I was reading it and I was like, wow, this is really well written. That must have been Demetrius.

00:11:43 Demetrius Malbrough And I'm saying the same thing about you, I said Rebecca must have written this part. So definitely, definitely. I love working with you. So we'll see what we can do in the future as far as, you know, really kind of bringing Puppet and Rubrik together, and having customers, you know, really request some of the things that both of our pieces of software does well.

00:12:04 Rebecca Fitzhugh Absolutely.

00:12:05 Demetrius Malbrough And also, I guess, have you heard of Puppet Bolt?

00:12:08 Rebecca Fitzhugh I have.

00:12:09 Demetrius Malbrough You have, and have you played around with it at all?

00:12:11 Rebecca Fitzhugh Yes, actually, funny story, I think it was maybe back in 2018. It was when I was one of vBrownBag kind of hosts, a co-host of a popular podcast, and I was really excited about Puppet Bolt and I actually helped coordinate a vBrownBag episode on that topic. And I used it a few times kind of around the end of 2018, around Christmas break. That seems to be when a lot of our lab work got done, if you recall. And I used it for managing a number of services running and some of our VMs, but I haven't used it much since then. And I need to. So are there any cool features that I should be aware of?

00:12:46 Demetrius Malbrough I'll probably have to get back with you on that. But definitely from what I hear is that it's really more like an ad hoc task base type of solution that you can actually you can install on your laptop and you can get started really, really quickly, really easily, and you can just kind of execute some of those things that you probably would have done manually. So take a look at that. And we also have Puppet Relay, which recently launched. Have you heard about that or played around with it at all?

00:13:13 Rebecca Fitzhugh You tipped me off on Puppet Relay a few weeks ago. You texted me like, have you looked at this? I hadn't looked at it until you texted me, and then I started digging into it and I liked what I saw. Our team manages a lot of our infrastructure using kind of infrastructure as code principles and GitOps, you know, as much as possible. And so I started playing around with it. And I haven't gotten too far, to be honest, but I started playing around with it to see how we can use it inside of our own customer's environment. And so you should know this Demetrius, I'm big on alerting. So my very first use case was kind of working to integrate it with our monitoring solution in order to generate JIRA tasks and then so I could assign them more easily to folks on the team. And so, so far it's been really easy to use and I like it and I'm hoping that we can use it more in our environment and that way we could come up with some more use cases, and I'd love to actually write about it.

00:14:02 Demetrius Malbrough So just a little more information to the listeners, is that it's our Puppet event-driven platform, and a lot of DevOps tools, a lot of DevOps teams can actually use this to automatically invoke pre-configured workflows. So it can definitely connect to a ton of DevOps tools like cloud services and APIs to automate some of the workflows across your platforms, so think pager duty and Git Hub and Data Dog and Jira, TerraForm and Slack. It's a pretty, pretty cool tool to be able to kind of do some triggered events. All right. One more question here for you, Rebecca. Do you have any fun ransomware stories that I may not know about that you would like to share?

00:14:42 Rebecca Fitzhugh Well, I would never use ransomware and fun in the same sentence unless we're high five-ing our customer because they are able to very easily recover. Yeah. So I think the one that I think of what kind of most recently was the city of Durham, so, side note for the listeners. Demetrius and I are both from the South, and as two Southerners, some of our favorite foods are chicken and waffles. So my favorite chicken and waffles place in North America is actually in Durham, North Carolina. So the city of Durham actually is a Rubrik customer and they detected ransomware. They had a ransomware attack on a Friday. That's horrible, and we keep seeing city governments, local governments keep getting attacked because, you know, so much of your emergency response systems are tied to local government. So how do you create mass havoc? You know, you really kind of wreak just pain across an environment, you attack a local government. Not that I'm trying to give anybody any ideas, but the mayor actually credited Rubrik for the very quick response. And so, like I mentioned on a Friday, they were attacked and everything was up, no problem, including things like even payroll by Monday morning the next week, because they were able to recover using Rubrik, because our backups are mutable, meaning that they cannot be consumed by ransomware, they can't be modified once that data is ingested. So really the key here's the mutability and we were really excited to hear the mayor actually directly name-check our products. That was really awesome.

00:16:14 Demetrius Malbrough Yeah, that was pretty awesome. I'm actually sitting here thinking that Puppet Relay could actually probably have something written that deals with ransomware as far as triggering, triggering an event and alerting. So I have to chew on that one a little bit.

00:16:31 Rebecca Fitzhugh Yeah. Maybe it's something like when radar alerts it pushes that to Relay and Relay can then push those alerts to other systems that you might be using.

00:16:41 Demetrius Malbrough We might be on to something. Let's see if we can jump on that lab of yours.

00:16:47 Rebecca Fitzhugh You probably still have access... Just kidding, you don't.

00:16:51 Demetrius Malbrough That would be horrible if I still had access. And plus, you probably don't want me playing around in there anymore because I am so rusty. It's ridiculous.

00:16:59 Rebecca Fitzhugh I think IT security would come after me if you still had access.

00:17:03 Demetrius Malbrough Yeah.

00:17:04 Rebecca Fitzhugh So if you have any questions, you know, feel free to reach out to me on Twitter, @RebeccaFitzhugh. Hopefully they spell my last name right in the show notes so that way we can get in contact. And if you're interested in more, kind of, automation goodness around Rubrik, you can follow at @RubrikBuild on Twitter as well.

00:17:19 Demetrius Malbrough And the spelling of her last name is F-I-T-Z-H-U-G-H.

00:17:24 Rebecca Fitzhugh Perfect.

00:17:25 Demetrius Malbrough Well, it's been amazing and fantastic to have a conversation with you again, Rebecca, and to talk things Puppet and Rubrik. I appreciate you coming on Pulling the Strings, and looking forward to doing a future episode with you also.

00:17:40 Rebecca Fitzhugh Absolutely. Maybe we can dig deeper into Relay later.

00:17:43 Demetrius Malbrough Yeah, we'll have to put that on the list. So thanks for coming on Pulling the Strings.

00:17:48 Rebecca Fitzhugh Absolutely. Thanks for having me.

Automate Service Level Agreements

Don't waste another hour doing something manually when it can be automated.