Product updates in Splunk-AppDynamics

Oh and welcome to today today sorry Welcome for today.00:00Day session. Today we're gonna cover Splunk AppDynamics enhancements and we're gonna cover releases 20 six.seven to 20 six.nine. My name's Kevin Morton. I'm a customer success specialist, and I've been with Cisco for around four years.00:15Today I'm joined by and Bill if you want to introduce yourselves. All right. Thank you Kevin. So I'm fired 90. I'm also have been with Cisco for around four four and a half years as a.00:32Observability CSS within Splunk customer success.00:52Looking forward to having this session today. Hi everyone, I'm. And I'm a customer success engineer previously a senior software engineering consultant. I'm with Cisco from six years now.00:57Okay, if we can go on to the next slide, please. Okay, so for for the 1st question, what we want to do is just get an idea of which platform you're currently using today, whether you're using SAS on premises or whether you're using the virtual appliance. If you are.01:15By using more than one you can select multiple.01:35I've got some virtual appliance users, that's good.01:44I'm very happy to see we have customers using our virtual appliance.02:11Yeah, definitely. Just give it a few more minutes cause I can see that there's still quite a few people to to let us know.02:15Okay, let's on to the next one. Okay, so in your day, day to day jobs, which of the Splunk app dynamics capabilities are you using are you using the.02:33APM are using end user monitoring, database visibility, infrastructure visibility, analytics or are you using it for, for application security as well. Again, it's multiple choice, so you can choose more than one if.02:51May mainly for for APM matt. That's brilliant, thank you. So if we're Oh, still going. Yeah ok that's brilliant, thank you. So if we can move move on to the next slide.03:20So for for the agenda today, we're gonna go through the control enhancements that came out in 20 six.seven. We're gonna cover the agent management enhancements, the agent enhancements, and the service services enhancements. And to kick this off, I'll start, I'll hand it over to.03:39Like I say, if you have got any Q and A questions that that you do want to raise, if you click onto the three dots in the bottom right hand corner, you should have an option for Slido and Q and A.04:01Thank you. Thanks Kevin. So as you know, the, our Splunk AppDynamics offering has a SaaS offering and the on premises offering in this session, we would like to cover actually both, but due to the interest of time, we just mentioned the highlights. Please refer to the to the documentation.04:11For the for the details. So with regards to our on premises offering, as you may know, we've released, we've modernized our enterprise console based on premises offering into a virtual appliance with the virtual appliance which is available for actually various.04:31Visualization platforms, you can facilitate, you know, installing, deploying and upgrading of your on premises infrastructure, and also you get the additional features such as.04:51Such as a secure app anomaly detection root cause analysis and lock observer connect which is an integration between AppDynamics on prem and Splunk. We actually have, we will have a dedicated EIS session on the virtual appliance in November.05:06So if you are interested please tune to that. And as I said in this session, we try to cover both platforms and we've indicated here on the top right of the slide to which, you know, offering platform the the feature, the slide, you know, associated is related to basically. So now let's.05:21Start with the health rule and alerts. So we have released the health rule suppression functionality which allows you to suppress health rules and the associated.05:41Events, actions during the win during the window. So this is this is good when you want to do e.g. a maintenance or if you want to suppress the, the, the health or violation for, for any reasons. So let me actually show you how to do that. You have to go to the other.05:54The response section and then to the actions and here we have action suppression, and you have to select the application and then.06:14Here we have two options. You have the one time kind of window option and the recurring option where you can actually specify the details and you have to give the scope by default, it actually will suppress all the health rules in your application. If you, if you want.06:23To, basically, only suppress subset of the health rules, you can also do that. And additionally, you can also disable the the reporting of the agents actually, so basically data collection reporting of of agents during that suppression window.06:43So this is actually available on both SaaS and and on prem environments now. Another update is due to basically the discontinuation of support of email to text notification which has been done by several carriers on our SaaS.07:03Platform it will be NO longer possible to send SMS alerts.07:23For those affected networks. So you can always use the HTTP request action to forward notification to Slack pager duty or other notification systems. Now I would like to hand over to Viva to continue with the licensing updates.07:29Thank you for it. So for the agent based licensing account where we now have the application level usage metrics and this is available also for the all users. You can see the usage unit and the usage count from the metric browser.07:48On the licensing section there is now space specific permission for the users, which is the license viewer. Basically, you will have to choose this role for the user that has to see the license UI.08:08And add it to the group or to the user specifically. Also on the licensing part, you can now see all the available all the licenses, all the license consumption, this grouped by.08:27The type of the monitoring that you are searching for, and also you can extract the result in a CSV file from the export usage button here.08:47On the licensing event page, you can now filter by the type of the events so we know we have seen a lot of errors reported from the agents like e.g. license limit exceeds not authorized, invalid access key and account license created expired.09:01Based on specific applications, so now you can see these events on the licensing UI and filter by specific events. Now let me show these three slides applied on our controller. So make sure that you.09:19You can see the licensing option here. You'll have the license summary reporting license rule account metric browser. So in the magic browser, you'll be able to see the.09:37Metrics group by applications only an application level. You choose the application that you want to see and then you have a class classification of grouping by the type of the agent. So it can be.net java, whatever.09:55Agent type that you have configured on this specific application. Inside you'll be able to see the usage, the license unit usage under reporting.10:11You are able now to select based on the typing of a specific name like just some part of it. You select the applications.10:30And you can see the license consumption. You can export the usage in a CSV file stored automatically in your computer. Then you have license rules, something that you have seen before was there from a long time. The account, we saw the metric browser, and now this is.10:46One of the best features we have regarding the events where you can filter based on the type of the event. So you see there are there are a lot of events similar like the health rules events that we have seen when health rules violated or the agent events, now we have this feature also for the licensing and now you can filter based on.11:07The license event. Let's suppose we have a license limit exceeded and then we select the application again we can type just part of the name what we remember, selected and apply and the filter applies. We don't have any events for this application.11:26Oh, let's go back to the powerpoint. So on this release, 26th or seven.11:45We add the model context protocol server integration and the MC MCP compatible AI clients can investigate read only monitoring data across APM, analytics events, infrastructure, Kubernetes database monitoring.12:04And the integration includes guided workflows, domain resources, list privilege access and HTTP and STDL transport options. What are the supported MCP transport? So the server supports two MCP transports, the.12:19The streamable HTTP for a shared HTTP HTTPS deployment and the STDL when an MCP client starts a server as a sub process. It uses two independent downstream credential sets, the AppDynamics controller authenticated with a request scoped barrier token for HTTP.12:35We are a process managed token.12:55Or an open authenticator credential for STDR. However, we have a lot of information regarding how MCP can be configured on our documentation side. Now, configuring the MCP server is very, very easy especially now that we have AI everywhere, and let me show you.12:57Now I did using cursor, but you can use any AI ID I chose to use cursor in this case. So I didn't even.13:17Did I didn't even do the configuration myself because I started doing like stupid easy prompts like configure for me MCP server for appdynamics. And then can you, yeah, can you try with demo one, which is our demo controller and give the steps according.13:26Then I was too lazy to do the configuration myself, so I asked the cursor to do it for me. It did the, it updated the configuration files for MCP.13:46After it so I did some troubleshooting thanks to the ID. It did the troubleshooting for me, then I was able to connect to the controller. I also sent some facts that I had the MCP API token configured correctly. I've already did the role that it requires.13:56So I was able to connect and then I asked which are the slowest business transactions for AD commerce for the last hour? So it listed me top slowest business transactions. So I noticed that the checkout is the slowest one and I did a missical.14:16Question what is wrong with the checkout business transaction and it explained to me what was wrong. Basically, this is gathering information from the API from the APM database and E one, so you have like a unified view of what is happening with this specific business transaction just by doing this simple prompt.14:36As I mentioned, you are free to use any AI ID to do the prompts and querying any, any, any question you have regarding the observability of your application.14:57On the agent agent management enhancement, we are now able to monitor who is changing like enabling and disabling the machine agent from the agent.15:23Management UI.15:36And this can be easily done using a simple, a simple curly command towards the controller, which will query the specific object for the, for this purpose. It, you can authenticate.15:37Using open authentic open authentication or you can use a specific API.15:57Now I am passing the ball to Farrid. So let's start with the agent management with the with the smart agent with version 26.29 here. The.16:05Sorry, I don't know why this yeah so with with smart agent version 20 six.nine, the upgrades, it's the agent updates is mostly about having a smoother and.16:23More reliable upgrade of the smart agent or the other agent with with the smart agent on windows. So for the smart agent self upgrade itself, now if the binary replacement fails on Windows.16:43The agent preserves its its configuration. When you want to upgrade the.net agent using the smart agent on Windows now there is a dedicated staging directory used to avoid conflicts, and also when it comes to a legacy package migration, this is done after agent.16:59Registration. So basically this all makes the the upgrade experience more smoother.17:18With regards to the java agent with version 20 six.seven, we've added support for for Kafka 4.1 and we also added support for play 3.0 with HTTP backends.17:26You also upgraded the log for J and Jackson data bind frameworks.17:45With version 20 six.nine, we've added support for spring bedflux 7.0 and .3 for play 3.3 .0. You can upgraded the Jackson data point library.17:49With this version 26.9, however, we did some security related changes that are actually breaking changes. So please pay attention to these breaking changes when you want to upgrade to the 02:20 six.nine, please also refer to the release notes for a completely.18:06For breaking changes. Here we mentioned two example changes. One is that by default now the agent will use HTTPS communication, you would have to use this agent property to enable HTTP and JMX ending browser right and invoke operations are disabled by default and you and you would have to.18:26Enabled them.18:46With regards to.net, we've upgraded the Splunk distribution of opentelemetry.net to version one point 15. So as you know, we've now introducing, we have introduced combined agents since since some time now where the idea here is that in your.18:49AppDynamics APM agent package there is not only the the appdynamics agent, but there is also a Splunk open telemetry agent, right? And this supports dual mode, which for.net agent you can enable by this parameter and in dual mode, what happens is that your applications is instrumented with appdynamics.19:07This agent and and also with this open telemetry agents, so you can have the telemetry data viewed both in AppDynamics and in Splunk observability cloud. This is for customers who want to test, you know, try open telemetry.19:27And in the open telemetry world, you have to configure some attributes such as service name space, service name and deployment environment name. The.net agent actually sets sets these attributes automatically for you.19:42With regards to note note.JS agents so with 20 with 20 six.seven .1 version, it was mostly about upgrading frameworks libraries, you know, and and removing dependencies. So Splunk Hotel agent was upgraded open SSL was upgraded and we removed.20:00I didn't see e.g. here.20:19With version 20 six.ten, we've added support for Node.js 26, and we actually removed support for older versions 18 to 21. Please note that since from version 20 six.ten now, we have a consolidated packaging right across.20:23Was all supported Node.js versions platform and architectures into a single distribution. So this might affect your automation tools if they still using the the earlier versioning formats. We've also removed some dependencies on this on this agent.20:43Now with that, I'd like to hand over Tomo to continue with the, with the machine agent. Be before going through the machine agent, because I mentioned the agent agent management enhancement and the API that is used.21:00Now to fetch the logs, I want to share with you some of the curl commands examples that you can use because I explained just theory probably doesn't make sense, but when you see the curl commands and feel free to do a print screen makes more sense to you. So this one is the the one that is using the user that you usually.21:15They use to sign in into AppDynamics and it's getting the include action object updated. Basically, it's very generic. Every object that is updated here will be on this audit report.21:35Then if you, you, you, if you want to use the authoriz authorization bearer like the API token, this is the command that you have to use. And if you want to filter down to a specific.21:48Objects, then you have to use this filtering here object type machine_instance, and this will filter only those object updated that are related with the machine instance. Regarding the.22:03And time, this is also important because in case you have to include the time zone, then the format changes. So this is just for G GMT, but there is another time format if you want to use and consider the time zone. Now, now I'm moving to the quickly to the.22:20Machine agent. So with the release 20 six.eight, the machine agent includes several key updates related to the docker monitoring remediation action log validation, open telemetry.22:41Regarding the collector life cycle management and configuration enhancements. Regarding the docker, the docker PID level process magic collection, which is now opt in, which means it is disabled by by default and must be explicitly enabled. The machine agent enforces the docker.22:57Individual process selector which filters which container processes are monitored based on the configurable criteria that you will define. Users can overwrite host and docker process selectors using environment variables or JVM properties allowing flexible.23:18Customization of which process and hosts are being monitored.23:38And these settings are configured in the docker visibility configuration files and relate to the machine agent settings for server visibility. Also this version allows, allows validation for remediation action log pulse.23:43So an action allow allow list validation feature has been added for remediation action log paths to enhance security and control and this validation is disabled by default to maintain backwards compatibility. You will have to enable it by asking your support team to enable the specific respective flag for it.23:59The bundled open telemetry collector collector lifecycle update updates for RPM installations. So for Linux RPM installations using system D, the life cycle management of the bundled auto collector has been.24:21Are now updated. Fresh installations leave the auto collector disabled and inactive by default.24:39Upgrades preserve the existing service state of the auto collector ensuring continu continuity of monitoring configurations. Also, the bandwidth auto collector configuration now includes a dynamics account resource attributes and this enhancements support combined agent function, functionality.24:4824.25:08Infrastructure visibility enable enabling better collaboration and telemetry reporting. We always do bug fixes and updates of the libraries. So this release also includes various bug fixes and updates to bundle components to improve stability.25:09Performance and these updates.25:29Improve the flexibility, security, observability capabilities of the machinery agent especially in containerized environments using Docker and open telemetry. For the class, for the cluster agent, the release 20 six.seven introduces not a.25:31Everything and monitoring. So when you click on the notes in the entity map pane, which is available only on.25:51Starting from 20 six.seven, it will list all the associated nodes to that cluster. So clicking a note from the list opens a dashboard that displays all events associated with a specific note, including.26:00Those generated by its underlying ports. So you can now view not health storage CPU, and memory metrics from the cluster agent UI on the controller. This release also adds support for Red Hat openshift container platform version four.26:17For the 20 and for.26:36Dot 21. Removes the requirement to configure enable vitometric and enable additional metrics flags, so you don't need to configure this because they are enabled by default and mark sensitive data import logs. Also there are upgrades of the Golang go language used to code to program.26:37From this agent.26:57And some bug fixes which you can check on the agent resolved issues on our documentation. For the database agent for the release 26 to seven, this remains at support to the Mongo database Atlas private endpoint discovery in the.26:59Database agent section when you configure the database collectors. So when you do the configuration of the database collector, you can enable the private endpoint discovery with a feature flag and then configure the Mongo database collector to use either the atlas host name or the private endpoint IP address and the port.27:19Based on the network access configuration, and this will discover the.27:39The Mongo database Atlas endpoint for the public endpoint host names, so if you want to monitor the Mongo database atlas cluster, add the IP address of the host running the database agent to the database and network access and this will make.27:43Possible the discovery of this public endpoint for for the cluster also you have the this flag here which is set to true if you want to use the private endpoint discovery.28:01When you, when you start the database agent, please add this flag when you want, if you want to do the discovery of the Mongo database private endpoint.28:20So for the Python agent 20 six.seven.one, we include, we have included the general availability for open telemetry combined mode, which means it supports open telemetry now and you can install the Splunk appdynamics package with this hotel extra.28:36You do pipe install appdynamics and you add hotel in square brackets.28:57Then you can enable the deployment mode to do all our hotel by setting the appdynamics agent deployment mode environment variable to do all our to single, to do all our hotel yes. Now it supports Python three point ten or higher.29:03And this is a requirement for open telemetry only modes and updated proto buff handling to use the CUPP implementation by default and to use this you'll have to enable the settings protocol buffer Python implementation equals to Python.29:23So the update to proto buff handling means that the Python agent now uses the C UBP implementation for protocol buffers by default, and this implementation is a faster, more efficient version written in C.29:42The previous available option is now deprecated and has NO effect meaning it should NO longer be used. So if you need to use the pure python implementation to protobuf instead of the default CUPB version, you can do so by setting the environment.29:59Variable that I mentioned. Again for the Python agent release 20 six.seven.two, this includes key, some key updates.30:16Like the support obtain shell free Java proxy startup for Debian GLIBC this containers. This can be enabled by setting the environment variable shell free proxy startup equals to true for the configuration proxy.30:32Shell free start top equals to true. So the default setting remains false. Regarding the this image containers your app for your application is or it's it's runtime dependencies.30:51And so the the this image containers has just your application runtime runtime dependencies and nothing else. No shell, not package manager NO cut none of the underland utilities that ship with the normal Linux space images like Demien or ubunto. So now we provide support for.31:08This as well.31:28The control communication now is modified to use TLS by default and ports 4431 NO port is explicitly configured and the release bundles the approved remediated 11.0 point 32 runtime.31:30Though without GRE packages remain unchanged, these updates enhance this security by enforcing DLS communication by default and improve compatibility with containerized environments by supporting shell free proxy startup. In addition.31:48The Python agent supports configuration through environment variables and configuration files including settings for TLS versions, 1.2 and one 1.3 proxy settings and reuse of node names for efficient monitoring in dynamic environments.32:09For the private synthetic agent, this release 26 to seven adds support for Selenium four point 43.0 and for more instructions how to install and configure it, please check on our documentation or contact us.32:30And on 20 six.ten, this release includes support for Chrome 153. The hosted agent, it supports Chrome version 151 from the release 20 six.seven.32:45And now back to thank you so it's about our mobile real user monitoring where we have the dedicated agents right for the different mobile devices. So for Android devices, we have the android agent with version 20 six.eight. We added support for.33:05For gradle plugin version 9.9 .1.33:25And with the iOS agent with version 20 six.ten we have added support for iOS platforms 15 or later and some bug fixes. And starting from this release 20 six.ten, we not only report metrics from the app delegate class but also.33:30From the UI scene delegate class.33:51Last but not least with regards to our open telemetry ingestion gateway, which is actually it's it's based on the open Splunk open telemetry collector and what the ingestion gateway is that it sits between your instrumented open telemet.33:56Telemetry application and AppDynamics controller, and it has a specific exporter where transforms open telemetry data into the AppDynamics data model and you can view the data in appdynamics. So with this, we've released 20 six.seven of the gateway, we've added self telemetry.34:15For the for this.34:35Gateway and we've added logs pipeline as configuration option. Finally, we would like to also give you some updates about our SaaS services. So our synthetic services we with that we, we, we.34:37Did some bug fixes and code optimizations. This is for the last months up to the version 20 six.seven.34:57For our end user monitoring and report services, we did, we did a library and 3rd party components upgrades. So these are our SaaS services, you know, running on our environment and makes it more secure. There is NO actions for you.35:05To take. And with that, we'd like to close the session and I hand over back to Kevin. Thank you. Okay, thank you for that. What what we'd like to do now is for you to just use Slido again to to give us.35:24Items that you'd be interested in future sessions to to cover, with, with that it helps us build out the the next expert in series and make make makes it more useful for, for yourselves.35:44So if people can just get go go back onto Slido and they'll have the the option to give us that.36:00This could be anything from like smart agents up to driver agent updates.36:29Yeah is that on pro processes on how to to update the Java agents or just.36:50There's new security features come for the Java agents just want to put that in chat.37:01Okay, that's.37:20Yeah, thank you. Move to the next slide. Okay, this this slide is just to make you aware of Cisco university. So if you, you do have an account, if you log into you.cisco.com, in there you can find courses and and short videos on.37:37On best practices around using Splunk appdynamics. And finally, we, we have got a YouTube channel, so if you do go to the Cisco observability.37:56In there you'll, you'll find short videos of how commonly used practices. If there's ones in there that you can't see if you let your customer success team know that then we can build, build one out and put it on there for to make available.38:12Next slide. Okay, and one clear final thought it is a big thank you for for your time. We we hope that this is this does.38:33It give you what you need and answers questions that, that you do have, and we do this to help you get the most out of AppDynamics.38:49