PurePerformance - Automated intelligence for PaaS platforms featuring Red Hat with Peter Hack

Episode Date: January 29, 2019

Peter Hack joins us to chat about how Dynatrace helps you to get better visibility into your PaaS environment; from deployment to automation. He shares a few tips from his deep dive session into OpenShift featuring Red Hat.

Coming to you from Dynatrace Perform in Las Vegas, it's Pure Performance! Hello from Dynatrace Perform 2019 in Las Vegas. I'm Andy Grabner and this is Up Close and Personal with another member of my team. I had Mike Villager already on the call. Oh that's awesome. Yeah well this time with Peter Heck. Welcome on the show Peter. You just came out of a breakout session that you did with I believe with Red Hat. Yeah we together with Chris Morgan we presented basically OpenShift, how Dynatrace can help to monitor OpenShift, automation within OpenShift and really to enable the things like Ansible and such. Cool, so

OpenShift, you know, hot topic, you know, as is obviously PaaS in general, Kubernetes, I know we are partnering, you are I believe the global technical partner evangelist lead for Red Hat? Exactly. So I work closely with Red Hat on all things OpenShift together with Dynatrace to really enable our customers and partners to be able to utilize OpenShift and Dynatrace as a combined effort. Cool. So tell me a little bit about the session. So what was the message? People that didn't make it to the session, what did they miss? Why should they watch the recording potentially?

Yeah, so great question Andy. So basically it was trying to understand for our customers why OpenShift is different, why it's a great alternative to just upstream Kubernetes. That it's enterprise-grade, it brings a lot of security to the mix. How Dynatrace closely integrates with OpenShift and the Kubernetes space, and is able to provide real observability into your OpenShift clusters and all of that nature. Also, because of the power of the Kubernetes platform

together with Dynatrace's automation, we were able to utilize products like Ansible, which is also a Red Hat product, to automate the remediation of problems that exist inside of those environments. like Ansible, which is also a Red Hat product, to automate the remediation of problems that exist inside of those environments. So we had a good discussion with our folks who joined us to really understand how they can really capture and utilize the value of OpenShift and Dynastrace together.

Pretty cool. So maybe let's start from the beginning with what you just said. I think the first thing you mentioned is our support for OpenShift monitoring. I believe we have the OpenShift operator, or the Kubernetes operator. What's the right? So the Kubernetes operator framework is basically the platform equivalent of a human operator

doing this. And so Dynatrace, together with Red Hat, produced a Red Hat Dynatrace together with Red Hat produced a Dynatrace one agent operator that can be run on Kubernetes or on the Red Hat OpenShift platform. And basically that operator makes sure that whatever I do on the platform is automatically monitored,

injecting the one agent fully automatically for me? Yeah, absolutely. So the concept of the one agent operator is that Dynatrace will deploy your one agent to all of the nodes within your cluster to easily be able to monitor them and to understand what's going on inside of your cluster. So you get all of that monitoring

as part of the platform. And together with that, we also capture and bring in additional OpenShift-type metrics about node names and clusters and projects and things like that. So there's a really good, tight integration there. And now you just mentioned Ansible, another product from Red Hat. So what were the talking points? What did you talk about in the session that is relevant to Ansible?

So there's actually another Ansible breakout that's happening in the partner theater. But what we did is we just touched on the ideas of how you could capture some value by utilizing a solution like Ansible to automate the remediation steps that you might have for problems that exist in your environment. Together with the operator, the operator can help to manage the life cycle

of the one agent and the applications. There are other vendors out there like Couchbase and those that are doing operators as well. So the idea is ultimately that together with these operators and products like Ansible, having a full automation lifecycle within your environment. Pretty cool. And this is just from my understanding. I know we talk about Ansible. I think there's also Ansible Tower. Is this what we're talking about when we talk about auto-remodulation or is it the same thing?

Yeah. So that's an excellent question. So Ansible Tower is actually the product that Red Hat sells, which is a single server that can provide all of the Ansible-related playbooks to do such remediation. So in our remediation model with Ansible, we actually use Ansible Tower and send any events or problem notifications to Ansible

for it to be able to act upon playbooks that can remediate specific problems within the Kubernetes cluster or elsewhere that we're observing. Ansible itself is just an open source solution so you can use Ansible, it's just scripts that you would run on a host. So that's really the difference. Ansible Tower is actually the product server that they provide. Okay, pretty cool. And can you give us some additional examples on auto-remediation other than, I don't know, maybe provisioning more hardware, cleaning disk space?

Yeah, right. The whole goal here is really to be far above the layer of just infrastructure remediation because Kubernetes handles a lot of the platform events like scale up, scale down based on resources. What Dynatrace brings to the table is really a capturing and an understanding of application level metrics, response times, things like that, so that when there's problems in the environment that are observed from an impact perspective, maybe users are being impacted by services downstream, Dynatrace could then utilize that information and provide those details to Ansible to be able to make very smart decisions about how we remediate.

So the intelligence of Dynatrace together with Davis can publish into Ansible specific playbooks that it should be executing, and those executions will then, for instance, scale up, scale down, maybe roll back or at least change your Istio config so that you're specifically you're now instead of being on blue which is problematic you can roll back over to your green deployment. Pretty cool, yeah. So obviously we've had a very tight partnership with Red Hat. You had Chris Morgan co-presenting with you.

Anything else that is worth mentioning from your breakout? Any other topics that you covered? Anything that people should be aware of? Yeah, I mean, I think overall the response was that a lot of the folks in the room are really driving towards OpenShift. They see the need for OpenShift in the space because the upstream Kubernetes is just not hardened enough today, and that's really what Red Hat is trying to bring to the table. And Dynatrace, together with that, everyone's starting to recognize the value. The automation story is really important.

I think there's a lot of opportunities to learn more about that here at Perform. But I believe that kind of giving a taste of that in the room, we got a really good response from the folks that were attending. Perfect. So I know Perform is still going on. So you said the partner theater? Yeah, the partner theater. So there's within our, I I guess within the conference, there's an area where you can go and learn more about Dynatrace and together with the product managers and such and then real close to that as well is an area where the partners are presenting specific solution presentations. So Ansible would be one of those I believe that Red Hat is going to be presenting about.

Perfect. So in case you happen to be in Vegas at Perform, visit us as long as the, I think the exhibition or the marketplace is open. In case you are not there, people that are not there, they're listening in, can you say what's the best way to get in touch with you? Because you are obviously the public face from Dynatrace when it comes to OpenShift with Red Hat. Sure. There's a number of ways to get in touch with me. You can either go through my email. I'm pretty easy.

It's peter.hack at dynatrace.com. You can also reach me through Twitter at hack at DT. So H-A-C-K-A-T-D-T. And I'll be quick to respond as best I can. Awesome. Perfect. Well, enjoy the rest of the show. We can hear in the background, not sure if the background noise makes it to the recording, but people are enjoying the break as well here. Let's have a drink probably or something, right?

Cheers. Yeah. So thanks Peter for a pure performance. I'm Andy Grabner.