New Splunk AppDynamics Features in v26​

1st of all, thank you all for taking time out of your day to attend this expert insight series session.00:10It's, let me just as you can see here, I'm sharing a slido. That's basically we have a couple of questions we're gonna ask within the deck for you and we would love you guys to participate and get your feedback. If you have any.00:17Questions, please just go to the Q and A tab and ask. I'll be coordinating today's session. So I'll be responding to any questions you may have during the session, you know, happy to ask any slides or for 8 % may resonate with you.00:32And you may have wish to have a greater understanding, please ask in the, in the Q and A chat and I will respond accordingly.00:52So the agenda today is actually, I think I went ahead of my let me see. Yeah, sorry. Apologies. The agenda today is Andrew.01:01Which is going to be presented by my colleague Farid and controller enhancements, agent management enhancement and agent enhancements. Just to reiterate, the topic of today's session is the Splunk AppDynamics enhancements from release 20 six.one.01:16And also agent release up to 20 six.three. So there's a lot of information you'll hear about today. It's really informative. It will take the form of both the PowerPoint slides and also.01:35Onto your controller to highlight, you know, and practice these specific enhancements just to give a better context to the theory on what they mean in practice. My two colleagues who will be presenting today's session is.01:51Balma, do you want me to do a quick introduction of yourself? Sure. Hi all, I'm Bill Mccaram and I am a customer success engineer from two years now at Cisco. I focused on observability, but not only. Before this, I was part of the delivery team as a senior software engineer consultant.02:06For other four years, part of Cisco again.02:26Thank you. Thanks very much. So hi everyone, I'm sorry I'm also customer success engineer in Splunk app dynamics within Cisco. I've been, I have been with Cisco for about four years before that was an APM consultant with more than ten years experience.02:30Fantastic. So guys, with those short introductions, you're in very good hands for today's session. So you can see here quick slide on the agenda, and we're gonna go to our 1st question on Slido, which is relates to basically we want to get from the audience.02:52What platform they're currently using in the context of Splunk appdynamics. Because I think it's very important to understand today's session is gonna cover all the different platforms. Some features will not be available in some platforms, but.03:10Will call that out. So that's why we wanted to, you know, try to get an understanding of the makeup of the audience and the context of what type of Splunk AppDynamics platform they're currently using. So we've got nine people on Slido responding and yeah, so you can see it's a mixture of their sad.03:28Has changed and as we, as we speak, we get more people joining the Slido as well. So this is great. Thank you very much. I'm happy to see that we have attendees that already have deployed the virtual appliance. That's great news. Yeah. Oh yeah, a hundred percent.03:47Oh yep it's great. And this just shows the flexibility of our, you know, current offering and that we support SaaS or on prem virtual appliance and then the classic on prem as well. Oh, this is good. We can sort of see a mixture sort of with just SaaS and the other on prem.04:04Solutions is that it's right down probably 50.04:24Actually it's gone over now it's more on prem. There we go, you've got more people joining as well. This is great. And this gives us feedback on what we would like to do going forward as well. There will be another Slido question at the end of this session. Remember the session is for 1 h and we plan to, you know, do a hard stop and.04:28On the, on the iron mark just because we all know you have your day jobs to do as well.04:48And alert calls to attend to. Right, we're gonna move on now because I know time is a ticking and I'm going to move next. It's gonna go to my colleague Varied, who's gonna do the 1st part, which is the intro setting the context around observable.04:54The SAS and also.05:14On prem as well. Pretty. Thank you Cherard, so, so let's start with a quick positioning of AppDynamics within the Splunk observability portfolio. So as you.05:16You know we rebranded AppDynamics to Splunk AppDynamics to emphasize that it is part of the Splunk observability portfolio. So within our portfolio, we have the Splunk platform. That's what customers use for enterprise log management and security use cases. We have appdynamics, which.05:36The proven APM solution formises. We have Splunk ITSI, which is our AI OPs platform and we have Splunk observability cloud, which is, you know, for cloud 1st solutions. Our strategy is to bring.05:55Unified observability experience across all these products.06:15Now when it comes to AppDynamics for our on premises customers, we released for, we have released for more than a year, a new deployment model, so the new deployment architecture called the app the virtual appliance. So traditionally for our on premises customers, you had to deploy.06:20By different components enterprise console, controller, event service, EUM server. So with the virtual appliance, we provide you with prebuilt virtual machine images that includes all these components.06:39So it makes it easier for you to manage the your on premises deployment. But not only that, with this with the virtual appliance, we bridge the gap and shorten the gap when it comes to feature parity between our SaaS and our on prem.06:53So some features that were traditionally SaaS only are now available on virtual appliance. So e.g., features like secure app anomaly detection, root cause analysis, log observer connected, we're going to look at today integration between AppDynamics and Splunk platform and open telemetry are now available.07:14As well on virtual appliance, so.07:33The features that we're going to look today, controller and agent and agent management features, most of them are available across all platforms with the excep with the exception of AI generated summary and, you know, Kubernetes alerting. So for these two they are not yet available on premises, they will come.07:36Later and we will point out also if there is some differences to the in the platform.07:56So with that, I'd like to hand over to my Colleaguema to start with the controller enhancements. Can you see my screen? Yeah. Okay, thank you. So from our last expert insight session, we saw the one click.08:03Root cause analysis for APM applied to business transaction databases servers, there's and nodes. Now from 20 six.one, this feature has expanded into the. And when I say, it's browser real user monitoring and mobile real real user monitoring. So now we have a.08:23AI driven troubleshooting for our web applications and mobile applications with insights for the base page experience through a respective metric and network requests for the mobile apps. Also we are.08:43Advancing more to to include more metrics into the AI driven troubleshooting. So the AppDynamics uses AI to correlate all the metrics associated with an anomaly and generates comprehensive summaries.08:57Of the issue and the top three suspected causes which helps you to understand the issue better. It ranks the top three suspected causes and provides recommendations on how to resolve.09:13The issue. So this saves you a lot of time, so if before you had a problem from the traditional anomaly detection, you would go to Google, Google for the issue and try to find possible solution. Now it's not anymore necessary because you have the recommended solutions there, which also are associated with evidence from the issue.09:28So you will have links to the snapshots that will link you directly to the flow map of the application or to the base page where you can check the issue yourself. Maybe this is also not necessary because having the AI generated summary.09:48And the recommended recommend recommended issue, you can directly implement the solution. So now let's demo this because it's easier to see this on action. So this is the one click.10:04RCA, the one click root cause analysis. You'll see an AI generated summary. I would say this is very human text.10:24Sounds like it's it has been written from human. Instead of the dry text you would see before from health rules summary or the traditional anomaly detection, then you'll see on the suspected root cause.10:35The three top causes and also on the right side, out of the issue, you'll see the last part recommendations. So you'll see what.10:52AppDynamics is recommending you to do to resolve the problem. How do I go here? So I go to the user experience, then I choose the.11:08You am the application that is having a problem, then if you want to have a detailed look of pages and Ajax requests that maybe are having any issues, you can do it but if you want, if you don't have time and you want to be fast, you would go directly to the alert and respond then to anomaly detection.11:23You'll see the model, the model is being trained and you check the base page that is having a problem, you click on it, you find a critical one, but if there is NO critical one, you, you try to find the.11:43Base page with an anomaly and you click on it and it will generate the AI summary root cause analysis with a recommendation.12:00Yeah. And you click on the suspected root causes. In this case there is just one. If there were more, you would see more root causes listed here based on the priority and then a description of the problem and recommendations of it. The beautiful part of it is.12:19This part, the evidence, so the resources, you click on it, you don't need to search for the evidence, you just click on the evidence and it will show you the basically the problem of the the the where the problem is on the base base page.12:38So as we have AI, we have applied it also on our health rules. So for a health rule violation related to a business transaction, you can now.13:00Now view the AI generated summary as I show you as I showed you for the anomaly detection. Now this applies also for the health health rule and you'll see also the suspected root cause. You can view the top three suspected root causes and the AI generated summary of the causes related to the health rule violation and.13:20As I said previously, this helps to reduce the mean time to identify performance issues in your application and infrastructure. I assume most of you remember how the health rule violation summary used to be before. So let's recall how troubleshooting with health rule violation used.13:41To work?14:00But we often were presented with a dry summary of numbers above the baseline or humanly specified thresholds. Engineers had to find the issue by researching between errors and deep diving into stack classes and methods. And this led to a long.14:01Rushing and cumbersome troubleshooting with NO intelligence expected root cause or summarizations. So now we have.14:21On AIs some generated summary, you don't have to go and search for snapshots or full graph snapshots, go through the call stack and so on because everything is here. Also with the evidence with snapshots linked here that can help you.14:29Understand more of the problem. If you have time, if you don't have time, go through the recommendations and you'll have the solutions there. Let me demo this so.14:45If I go to an application and select.15:04One, I see here that I have health rule violations, a hundred 203a lot to have for an example, then I let's.15:08I have a problem with the checkout.15:28Yeah Usually it's faster than this, but keep in mind this is our demo environment and a lot of things happen in the background like upgrades, people using it and so on. So you see the AI generated summary.15:36It's also giving you the numbers as the traditional one was doing, but also it describes how the violation has changed over time, basically the timeline and gives you the suspected root cause, so seems like.15:53This is the database performance issue and it's saying that this database has a problem with the queries and the complex query.16:12Especially on the joins, basically there could have been wrong joins used in the query and so on. And you have the snapshots listed here to see, to get more details about the issue.16:27From 20 six.one.16:47When you create health rules for the mobile applications, now you can use the metrics that we used to see on the out of the box dashboards as part of the metric browser now in your health rules. So now you you're able to choose the network request on the metric browser.16:50And create health rules on them. So this is a very nice feature. This is, this gives you the possibility to use the relative path of a metric so you can generalize it more for a health rule. And I'll give you a demo after.17:07But not apply to health rule but to a dashboard because giving like demonstrating demonstrating it from a visual perspective makes it easier to understand also from a health rule perspective. So this updates.17:27Introduces more precise control when defining health rules using relative metric pass. You can now enforce stricter matching with the exact match, one wildcard on one directory level option.17:40So previously wildcards would match multiple levels, which sometimes led to overly broad or noisy evolution evaluations. So with these enhancements, each wildcard maps to exactly one segment in the metric path, and this ensures the health rule evaluates only the intended metric.17:55And avoids unintended matches. So I have an example here like custom metrics, webspare, a queue.18:15A Hyphen two star and status, so this will match only this metric but won't match the others because we have selected the exact match on one wildcard to one level of directory. Now.18:24I don't want to be very abstract I I will showcase this on the demos through a widget to understand this better.18:41So you see there is a list of, let me go back. So basically I'm monitoring here the memory, but we have two types of memories HIP and non hip. And I want them both in the same graph. What I do is.19:20I use a relative path and I replace the type of the memory with asterisks, and this will give me memory and heap and non hip memory both in one single chart. The same practice, the same idea is also applied.19:37When configuring a health, a health rule.19:57From 20 six.one, you can now monitor the historical notes by filtering historical tiers and notes on the metric browser browser but also on the UI, this gives you the possibility and the visibility on the notes that have been marked as historical and remo.20:08Moved from your UI because you know when when all these markets historically, it's removed from the UI and you have NO possibility to.20:28To monitor this. But from 20 six.one if you want to monitor those and also you want to create health rules or have a specific dashboard just for the historical notes. Now you have this specific filter applied to the historical notes and another demo.20:37So now on the tiers and notes, you can filter and you can add the criteria of historical notes and this is 1 h basically. You'll see all the agents are down from for the last 1 h because these are marked historical from 20 days ago. So if I want to have more details to.20:56See metrics.21:16About these agents, they are all reporting zero metrics. However, this is the filter, this is how you can select it and view the historical notes. Also from this version, you have the possibility to have to change the language of the control.21:17At the moment there are just two languages English and English and Japanese but we plan to add more. And as always, we keep our libraries.21:38To our to the last version to be compliant, security compliant and as performant as possible.21:48Progress on licensing is always one of the main responsibilities that we have as part of the as part of our product team. A lot of advancements and enhancements have been done to the licensing UI. So the license.22:03Page for both agent based licensing and infrastructure based licensing accounts is redesigned to deliver an improved user experience. So on the license rules page, now you can view the license usage details across all the applications with a single click on any license rule, and this minimizes the number.22:23Number of clicks required to access detailed license usage information. You can see the access key associated with your license rules. This was a feature before, but now you can see it for every rule. The license rules page now loads.22:43Faster than before, even if you have a lot of rules configured in in it. Also on the license summary page, you can now view the exact license usage count such as a thousand instead of one K So I I have written a thousand but it can be.22:57A full number, so you will not see the one K, the K after the the number, but the exact, the precise number of the license usage. You'll see the license expiry dates for all your license entitlements on the reporting page for the agent based licensing account.23:17You'll see the select application wizard, which is now loads the applications more quickly when you begin trying typing an application name and this allows you to find and select applications with improved speed and deficiency, and you can sort the columns in the usage breakdown table.23:34This section enables you to view applications according to the license usage and so on.23:53Also a new metric has been introduced license usage which is now published on the metric browser and then can be used part of your health rules to monitor the license usage or to to alert you before the limit that you have set is reached.24:00Back to our demo environment, so you are here and you go to license, you'll see the 1st license summary. This is a summary of all the type of agents that you're using for the application performance.24:17Then you have the user experience license split by synthetic monitoring and connected devices. Also you have the session replay usage, this comes from 25 to ten.24:37You have the database license usage, infrastructure visibility analytics secure application, and the data storage. So if you want to see how much data has been ingested from your controller, you have visibility in this part of the licensing summary. Then if you go to reporting, this is what I was saying before you.24:54Can quickly filter for a specific application.25:14Like I will filter for applications that contain commerce in the body and I'll select the one that I'm interested and I'll see the license usage. And you see the usage breakdown that I mentioned before and.25:18That you can use to order by the usage of the license for application name. You always can export the usage data in a CSV file without having the need to use any api's as before or having the need to have API accessibility and rights to access.25:35The official APIs that AppDynamics uses to export entities, one of them licenses. Then on the license.25:55Rules you have all the rules created. You can click on one and see the conception per type of agent. You can show the access key for every license rule. And as you see.26:04There is a long list of license rules and the rendering time was very fast. Then on the metric browser, as I mentioned, you have the license usage per license rule, which is split per per application then per type of agent. You can use these metrics here.26:20To create your own health rules and get alerts based on the usage of the license.26:40So you take your own measurements before you reach the limit and agents will not like the matches will not be ingested from the controller to prevent these cases, create these health rules and you'll be fine. Now I leave you with on the capable.26:47And so far it to.27:07Regarding dashboards. Thank you will know so let's now look at some of the capabilities we added to our dashboarding. So as you know, we have the classic or custom dashboards and we have-studio. So when it comes to classic dash.27:09Dashboards we added.27:29We added import button inside the the dashboard. We also improve had improvements when you want to share a dashboard and also visualization of, you know, color based visualization for.27:36For metric value widgets, so let's look at them in the controller yeah so we used to have when it comes to importing dashboards, we used we we have an import optionality option here on the dashboard. One when you do this.27:50A copy of an existing dashboard is created, right? So if assume you you have an existing dashboard, you export it, you want to reimport it, you do some modification and you you want to reimport it, and when you use the import button here, a copy of the existing dashboard is created. In this case, if you want to remove the previous dashboard, you would have to, you know, have the.28:09Need permissions for that user. What we did is that we added this import button here. If you use this now, the existing dashboard is update updated. So NO, new dashboard is created. It makes it easier, you know, for managing the dashboard.28:29Another feature we added is it it's with regards to sharing the dashboard. When you, when you share a dashboard, we added a full screen button here that when you use it, it removes the browser panel and, you know, it makes it more suitable to be displayed.28:48In in large monitors and so on that some of our customers are using.29:08Now going back to the dashboard, talking about the metric widget. So we have this metric value widget here and let me plot a sample metric.29:15E.g., errors yeah so we can compare the metric value with the baseline and then apply baseline color. So that by default, you know, if it's above the changes color.29:32We can even reverse that. So, if you reverse it then it becomes, you know, green unless it's, you know, this has the same value like here. So let me change the baseline just to have.29:52In different value. Yeah, so you see now it's below, so it will be green if you reverse it and otherwise it will be sorry it will be red if you reverse it and now if it will be green and you know it makes it easier than to visualize this.30:10So this is with regards to our improvements to, to custom dashboards. When it comes to-studio, as you know, this is what we we released-studio a couple of years ago and we are bridging also the gap with to with regards to function functionality between these two.30:26These types, two types of dashboards. For-studio, the interesting thing is, is the usage of variables, right? So in-studio, once you create a dashboard, you can actually have variables.30:46And this is particularly useful, you know, if you have, you know, you want to monitor different applications and so on with custom dashboards you you might have to create copies of the same dashboard for the different applications. With that studio, you can actually define variables, you can also define nested variables and have the same, you know, dashboard.31:01Template and just change the variable. You can even change the change it dynamically, right? When the user is viewing the dashboard.31:21And what we are doing is we are, you know, adding this support of variables, you know, to more, more and more widget types. Now moving on to the next topic cluster monitoring, right? So.31:28This is basically our Kubernetes monitoring capability based on the cluster agent. So what we've added to the controller is alerting on Kubernetes entities. This is something that has been requested by by several of our customers. So now.31:45Now if you go to the alerting section we can you can actually define alerts for Kubernetes entities. So this is behind a feature flag and we actually did an update on this, on this demo controller. So unfortunately the the feature had.32:05Has been disabled again so I can't show it here, but basically you have this Kubernetes option here.32:25And with that, we can have, we can have, we have access to metrics collected from our cluster and then we can use those metrics to define health rules. So we have, we can define health rules for.32:33You know, at the cluster level, at the workload level, at the name space level and even at the pod level. So for all these types, we, we can define the health rules and I also we So we will reference you to the to the documentation where you can actually list the.32:50Need a list of different metrics on which you can, you can define the health rules. This functionality is actually at the moment SAS only and it will come to the virtual appliance. The tentative date is 26th of seven, so in, in several months it should be coming there.33:10Moving on to the next topic, when we talked about our, you know, observability portfolio, we talked about, you know, having this unified observability experience and that's what we have been working on. When it comes to AppDynamics and Splunk platform, we have released the log observer connect. We released it.33:31More than a year ago and we have been enhancing this, this, this integration, so.33:49When you think about these products, we have AppDynamics, SaaS and on prem, and Splunk platform has also SaaS and on prem. So SaaS is called Splunk Cloud and the on prem is called Splunk enterprise. So you can set up the integration between all the variations. And here the enhancements is for the SaaS version if you have.33:55Of AppDynamics SaaS and Splunk Cloud. Now more, more regions are supported, you know, so basically the before it used to be the case that AppDynamics has and the Splunk Cloud used to be in the same region, but now it's.34:15It can be in different region and the integration can be set up. For our on premise customers, you need to have the visual appliance to set up this integration and then you can set it up with the, with both, you know, cloud and enterprise. And, I like to demo the, the, the integration.34:29To you on our SAS control. So if you go to to an application, once you set up this integration, we have this smart button.34:49New logs in Splunk that will be enabled. And once you click on that, it brings the user into Splunk in this case Splunk Cloud and it shows the user logs. So contextually relevant logs.35:02From where the user, you know, made the click. So we were, we made the click on an application level, so we we can see all the logs for that application and in in this case the interesting thing with the integration is that it fields the, the, the, the query for you, right? So you don't have to worry about the, the, the.35:17Yeah.35:37The the query. And this button is available in various places such as, you know, tiers and notes and so on. But most interestingly, you know, in transaction snapshot, that's these are you know objects that we use a lot when we do troubleshooting. So this button is available also in the.35:38Transaction snapshot and it brings you to Splunk showing you, you know, based on the.35:57Geo ID of the transaction, so the logs are related to that specific business transaction. So in this case only seven logs. So this is showing you the contextual relevant logs, you know, for the for users who are you know comfortable, you know, doing more troubleshooting on the logs side.36:04On on on Splunk. Now this finishes our enhancements on the controller. Moving on to our agent management enhancements. This is the capability, you know, set of capabilities we've been working on.36:23Since couple of years to allow our customers to manage and deploy agents at scale easier.36:44So with agent management and this is all for all the platforms, right? So with the agent management we we released the API support for agent management to not only do the activities from GUI but also through API. We support now data deployment groups for.36:51Database agents and you can also now export data for our repository of agents from the GUI.37:11So let me show you this in the, in the controller. So if you go here, you have the agent management functionality, right? So this is the new functionality since couple of releases, and you have the repository of agents here and for each repository of agent type, you can.37:17And export the data.37:37And as you know, agent management to, to fully enjoy it, you you need smart agents, so once you install smart agent on on the VM or on the server, that basically server becomes managed, so you can perform activities such as updating the.37:39Agents or you know installing the agents, all of that through, through GUI, right? So.37:59This is the agent management functionality and as I said, with the support of API, you can now programmatically do this functionalities from the API. Now talking about deployment groups, so the idea of deployment groups, it's, it's managing group of agents.38:04You know, as a group, right? So we added supports of database agents, so you can specify.38:24The source where you want to, you know, install, e.g., update, get the newer version of the database agent. And then basically you specify the the properties and you select the host, right? And in this way you have a group of, database agents across different hosts and you can manage them.38:32Like this. So, yeah, so this is about our agent management capabilities and moving on to our agent enhancements so we we have a lot of I mean there's a lot of enhancements to to talk about, you know, as you know, we regularly update the frameworks of.38:52Our agents, you know, when there are security vulnerabilities and so on and you know reduce the remove the deprecated ones. We due to interest of time, we selected subset of the features.39:11With regards to smart agent. So this is, as I said, allows you really to fully enjoy agent management capabilities. When it comes to deploying a smart agent, you know, at scale, we have the, it has the command line interface for doing that and we are adding more functionalities there.39:22You know when it comes to deploying this at scale through the command line interface, e.g..39:42SSH password authentication and so on. Continuing on to the cluster agent. So this is you know part of our Kubernetes monitoring. So, it's kind of both, right? We have improvements on the GUI and also on the agent.39:48This is at the moment, as I said unfortunately only for the SaaS customers for on prem it will be individual appliance. If you just for the on prem ones, you just enjoy, you know, the, the newer basically go, framework, so called go language, right? Which is the basis of the cluster agent. But on.40:08The on the GUI side with this new.40:28Cluster agent and the new new versions of the controller. We have not only the the alerting capabilities but also improvement to the GUI, right? So, going to that, to the graphical user interface, going to the servers, to the clusters here.40:32You you can view the the cluster. These are the.40:51Default dashboards for monitoring our, our clusters, our Kubernetes clusters so that what has been added and again unfortunately I cannot show you because of the feature flag is that we, we did improvements in the in the UI, so basically we added these entity maps.40:55Here on the left hand side.41:15So, you, it makes it easier and more efficient to navigate in the, in the UI, right? When it comes to navigating through different entities, so that basically one enhancement alerting I talked about, and you know we have more native monitoring capabilities, right? With, with the with the metrics that are.41:16Being collected for the different entities, name space, workflows, the workload parts and cluster, you can you can.41:36We have we have a much more better monitoring at, at this level. Moving on to the next next type of agent, the not the.net agent. So what we did here is that we now support combined agent supports now a.net agent.41:44Agent so let me explain what is the idea behind the the combined agent. So.42:04For those customers who are adopting open telemetry or or are thinking about adopting open telemetry, testing it, the combined concept of combined agent agent makes it very easy for them to test this. So with the combined agent, you have basically it's the same agent, the application server agent, it has.42:09The app the server agent inside which sends telemetry data through appdynamics, right? Whether SaaS, whether on prem, but it also has another component, the Splunk open telemetry agent, which can once activated.42:28Can send telemetry data in open telemetry format to your open telemetry backend of choice. It can be AppDynamics on premises, the virtual appliance supports now open telemetry. It can be, you know, appdynamics as it can be cloud or other open telemetry back end.42:43It allows the applications teams to try, you know, test basically what they get with open telemetry data. And now as you can see, it's, we supported already few types of agents such as Java and now we also support.net.43:03So with that, I'd like to hand over Tomo to continue with the agent agent enhancements. You're mute Willma.43:18Thanks. So from 20 six.one regarding the enhancements to the machine agent on the controller-info.XML, you know, it's the file that we use to configure our agents to point to the right controller and additional configurations. Now we have added more parameters.43:37Like the proxy configurations, proxy host port, user and proxy password file. So this can be done through the controller-info.XML which is part of the agent folder that you download from the portal of our agents. There are a lot of bug fixes on this version.43:56And also, there is a replacement of Netti with HTTP and Junix socket and also further 3rd party upgrades. So we know Netti is a powerful asynchronous event driven network framework, but it's heavy for search.44:16Some use cases where a simpler, more modern HTTP clients can be sufficient. So we replaced it with oak HTTP.44:36Which is the efficient HTTP client for Java and is designed to be faster and more lightweight. And by switching to ok HTTP, the machine agent achieves better performance when making HTTP calls reduces its memory footprint and simplifies the code base making it easier to maintain and very secure.44:46So we are not compromising the security when we do this replacement. When it comes to Unix Unix socket, this library allows job applications to communicate using Unix domain sockets UDS instead of the traditional TCP IP connections.45:05So why I'm emphasizing this? Why this matters? This matters because when the machine agent communicates with local services on the same host using TCP IP, this requires going through the full network stack, which adds latency and requires network port management.45:25Unix domain sockets UDC allows for faster, more secure communication because the data never leaves the local operating system kernel. And this is best practice for local inter process communication. So for the 3rd party.45:45Library upgrades we always upgrade them and we aim security patches like fixing C VS common vulnerabilities and exposures, bug fixes and compatibility ensuring that the agent works with newer version.46:04Options of Java and operating systems.46:24As mentioned about the combined agents for agents for the.net, it also applies to the machine agents so from 26 to three, the machine agent supports the combined agent for infrastru infrastructure visibility in which enables dual dual signal infrastructure monitoring.46:27And reporting telemetry.46:46To both AppDynamics and Splunk observability cloud through otel. And also this version of the agent includes following more 3rd party libraries upgrades and back fixes.46:50The database agent, this is one of the best improvements that we have done related to the database agent regarding the security. So this release includes the Microsoft intra ID authentication to connect to Azure databases.47:07For postgres SQL. So as you know, the database agent is responsible for collecting performance metrics from databases and to do this it must authenticate with the database instance. So historically this often required storing the database specific credentials like username.47:28Name password within the agent's configuration, but by adding.47:47The support for Microsoft entry ID, which is formerly Azure Active Directory authentication. The agent can now leverage token based authentication provided by the Azure ecosystem. So this enhances security by removing the need to manage.47:52And rotate static database password within the agent configuration, so changing this manually, and this aligns with the modern zero trust security models. So a short explanation about what is the Microsoft entry ID authentication. It's token based.48:10So instead of static password, you use a token, it's centralized security. So the access is managed centrally within the Azure portal and also ensures enhanced security. And what are the benefits of using this? So our customers can now meet stricter security and compliance requirements while maintaining full visibility.48:29Into the Postgres database performance. I will demo this later. The configuration is very, very easy, it's even funny I would say. So in addition to this, now the database agent can now publish up to 500000 metrics to support high number of metrics for database file utilizations and also.48:49Support the use of any cyber arc version in addition to hashicorp and CSM, so now we have this integration. We're working further to improve this integration with the dynamic tokens for both hashicorp and CSM.49:09Also, there are upgrades on the Microsoft SQL driver to keep it compatible with the security standards, which also will require an upgrade to the Java eleven to 17.49:25Regarding the configuration now for the J JDBC, you, by default it comes with the encryption to true, if you want to change it, you have to set the encryption to false, but by default it's.49:41The the JDBC connection is encrypted. You can change it though, and the explanation is on this slide. Now let me demo quickly.49:56So I have chosen Microsoft SQL server here.50:16And you'll see the possibility to add the database credentials using username and password cyberark and actually when you don't set a password here, it will automatically fetch the token from Microsoft entry ID. But to have this working, you'll have 1st to have an entry ID configured.50:20And.50:41The database agent will be able to fetch these tokens the private synthetic agent is part of our synthetic monitoring. So from.50:4320 six.one, the improvement that we have done related to to this agent is that now it can read parameters from a configuration file which makes it easier to manage complex installations.51:03Now it supports Chrome browser version 143 and you can choose it on the dropdown list when you make the configuration and also can be deployed on Kubernetes. The JavaScript agent is the job the script, the agent that is used for browser real user monitoring. It.51:17Used to capture the data from your application from your browser application and report it to the controller in the user experience part. Now it supports also a GraphQL request and it adds the last part.51:36It appends the last part on the request which is the operation name. So you will not see any more API/GraphQL. You also see an appended part of it, which is the operation name of the GraphQL. And to do this, you'll have to add some headers on the JavaScript snippet when you do the.51:56Configuration. Everything is well very well detailed on our official documentation, which is.52:16Public and you can access it. So basically these are three headers that have to be added on the snippet to enable GraphQL monitoring. And with this, we are at the end of it. Let Gerald share.52:22Thank you. Can everyone see the active poller? Yes, I can see it. Great, thank you Palma I just wanted to some things it doesn't work Webex. So folks, thank you for staying on. We literally just a couple of minutes to the oire. Thank you Palma and Farid for a lot of.52:39Information. What we're saying in the Slido here is basically we want to get your feedback on what topics you would like us to cover in the future because we're gonna keep on, keep doing these expert inside series sessions. We're getting a lot of positive feedback from our audience. We had a fan fantastic turnout today.52:59And I want to mention as well, there was a question in the Q and A This session is recorded and it will be shared the recording will be shared as well as the deck. I think that's very important. Why? Because there was a lot of information that Alma and presented.53:18On today and we're conscious that it's hard to take everything in with so much information so the deck will be shared for you guys to reference accordingly. So thank you. I see there's a lot of people on the poll in Slido providing their, their feedback. The ceilers are a lot around alert and respond, so we'll take that on board.53:38We'll create a session around that.53:58And we'll look at schedule and over the next few months as well. So keep your ears out on that. You'll get an email highlighting what's the next couple of sessions coming up over the next three or four months. But yeah, keep on providing that feedback. I just want to make sure we stop on the air because we.54:00You know where you're very, very busy.54:20I just want to call out Cisco U it's a Cisco University. It offers a lot of self paced and instructor led courses, gives you a deeper understanding around the capabilities of the Splunk AppDynamics platform. We're conscious this is 1 h session, so if you want to do a deeper dive into specific.54:23Areas, that's a good place to start. On on top of that, and this is what I actually love this new repository we've created on YouTube is the youTube.com at Cisco observable like a lot of very ingestible content.54:43Some can be three or 4 min in length, so it doesn't take up a lot of time around specific areas of the platform and then enhance your understanding like of our capabilities around that observability in the context of Splunk AppDynamics.54:59And honestly, I love this YouTube channel and a lot of customers who've come back to me have, you know, given really positive feedback as well. So just wanted to highlight that in case you're not aware of it. So again, something to reference, if you want to do a deeper understanding, a lot of content regularly.55:15Updated and added to that channel too as we speak. Okay, so.55:35Thank you, everyone, I'm gonna stop record stop sharing there. We'll stop recording as well, and we're just literally a minute past the hour.55:42