Probably the last area for me to touch on in this series of blog posts about the 18.104.22.168 Enterprise Manager release is the middleware management enhancements. That’s by no means because middleware management is the least important areas of enhancement, but rather simply a reflection of the fact it’s the area I know least about. It was a lot easier for me to bash out the other blog posts because my background has been mostly in the database arena. 🙂 Be that as it may, one of the best parts of working with the broader team I work with at Oracle Corporation is I have access to some of the best people in the business that can explain stuff like this to simple people like me. 😉
So what are the major middleware management enhancements in the 22.214.171.124 release? Well, there are quite a few!
- Diagnostic Pre-Check for Fusion Middleware (FMW)
- Target Management (awful term alert) Best Practices (/awful term alert) for FMW
- Enhanced Middleware Summary Page
- JVM Diagnostics as a Service
- Java Flight Recorder Integration
- More Administration Options
- First-Time Provisioning of FMW
- Manage Standalone Oracle HTTP Server 12c
- Enhancements to Coherence Management
See? Told you there was a lot! 🙂
Diagnostic Pre-Check for Fusion Middleware (FMW)
A lot of the time when people have issues when discovering FMW, they have problems trying to diagnose the root cause of why they can’t discover a particular domain or why a certain feature isn’t working the way they want it to. To address this in the 126.96.36.199 release, we have packaged up a series of tests that can be run either before discovery is performed or post-discovery. What these tests are going to do is check your environment to make sure that it is on par with a successful discovery. For example, if you have a domain that is not JRF enabled, the check will determine a JRF template has not been applied to the domain, which in turn means you won’t be able to use our Log File Viewer feature in Enterprise Manager Cloud Control. Tests like this will tell you when you won’t be able to perform discovery or enable particular features, and then provide a recommended course of action to address the problem, as shown in the screenshot below:
Target Management (awful term alert) Best Practices (/awful term alert) for FMW
After you have discovered your Fusion Middleware software, you may be wondering what’s the next thing to do – what features can I use now, how do I enable them and so on. That’s where Target Management Best Practices for FMW comes in. Once you’ve performed discovery, these recommendations are going to suggest the next steps for you to take. Basically, they are going to provide links to areas that we think you should go to after discovery. The first of these is to run the diagnostic pre-checks just to make sure everything is OK. Next we’ll guide you through the process of deploying the JVMD agents to collect detailed data on JVM threads, stacks, heap and CPU usage. The next item allows you to create and apply monitoring templates before finally defining compliance standards and associating those with targets:
Enhanced Middleware Summary Page
The Middleware Summary page has been redesigned and enhanced quite a bit with the 188.8.131.52 release to improve its performance and filtering capability, as well as to increase the visual appeal of the page. It now has a similar design to the All Targets page. By default, everything is displayed in a collapsed, hierarchical view, as you can see below:
You can, of course, expand any of these attributes, but you can also filter the data using the Search field or the pie charts, as well as by the target type, status or other target properties. The filters you use most recently are saved for the current session, as well as across sessions for the same login. You can also add a number of other pie charts to the summary region that might be of more value to you.
JVM Diagnostics as a Service
In the past, we’ve heard from customers who want to expose more of the diagnostics capabilities to more of their users, such as their application developers to help them diagnose problems. To address this, we’ve introduced a concept called JVM Diagnostics as a Service (a.k.a. JVMDaaS) which allows administrators to provide a self-service way for users such as application developers to easily gain access to JVM Diagnostics features. Essentially, they access a self-service application to download the diagnostics agents and deploy those on their desired JVM. This process doesn’t require anyone to perform discovery on the user’s JVMs. To enable this capability, you set a quota in terms of the number of JVMs being monitored.
Java Flight Recorder Integration
As I’ll discuss below, we’ve also integrated more of the WebLogic Admin console capabilities in the 184.108.40.206 release. We’re also integrating some of the Mission Control capabilities into the Cloud Control console, so we can now invoke a Java Flight Recorder (JFR) snapshot or dump directly from the Cloud Control console. You can also configure invocation of a JFR dump as a corrective action, so if you have a JVMD metric threshold set and the threshold is violated , a JFR dump will be automatically made.
From within Cloud Control itself, you can view a list of all the snapshots that have been taken and delete those you no longer need. To view a dump, you can use the Mission Control interface on your desktop, as shown below:
Currently with the R4 release, we support JRockit JVMs only. This will undoubtedly expand to other JVMs as time goes on.
More Administration Options
A couple of releases ago, we started integrating administration operation for WebLogic into Cloud Control, like the system MBean browser, JDBC datasource management and so on. That meant you didn’t need to drill down to the FMW Control Console or the admin console as much. In the 220.127.116.11 release, we continue to expand the configuration elements you are allowed to modify for WebLogic via Cloud Control. Specifically, you can view and edit settings for the domain, cluster, server, server template and machine configurations. All of these settings are integrated with the Change Center, so just like with the WebLogic console you can enable your configuration lock, make changes to the configuration, and then activate those changes:
First-Time Provisioning of FMW
In earlier releases, we provided support for cloning Oracle Homes or WebLogic Domains, but these cloning operations required that the Oracle Homes or WebLogic Domains had been already installed, discovered and being monitored with Cloud Control. Now you can use EM to perform first-time provisioning, which is accomplished by creating a provisioning profile. You download the JAR files from Oracle (via OTN or eDelivery) to your machine and then use Cloud Control to create this provisioning profile based on that (essentially, loading the JAR file into the software library). You can use this for WebLogic Server, SOA Suite and Oracle Service Bus. You can then launch a deployment procedure and point to the provisioning profile in order to do the first-time provisioning, using the UI to define the various installation directories, configure the domain (i.e. the number of servers, the ports and so on). You can also now front your domain with Oracle HTTP Server during this first-time provisioning, as well as during cloning or scaling out a domain. Of course, this will require that you have Oracle HTTP Server discovered and installed in order for us to do this configuration operation.
Manage Standalone Oracle HTTP Server 12c
Early on, about the only metric we supported for standalone Oracle HTTP Server was the response metric – is the standalone Oracle HTTP Server up or down. That was extended in the last release for standalone Oracle HTTP Server 11g to include much more performance monitoring, configuration management and process control features. In this latest release, we have added the same level of support for standalone OHS 12c, so you can centrally manage the web tier alongside application and data tiers. This includes:
- Out of the box performance metrics, spanning status, server (e.g. Active HTTP Requests, Request Process Time), resource usage, virtual host, response codes, and modules
- Automated collection of configuration data, including configuration files like httpd.conf
- Analysis of configuration data via comparisons, change tracking and search functionality
- The ability to view log messages and files
- The ability to perform start, stop and restart operations
Enhancements to Coherence Management
Finally, we’ve made substantial changes to our Coherence management. We’ve completely redesigned the user interface so it’s now ADF-based, and now includes the left hand target navigation, as well as regions for top components across all caches and all nodes, as shown below:
As you can see from all this, there have been LOTS of improvements in the 18.104.22.168 release for middleware management.