VMware Tanzu Application Service 2.9: Key Enhancements for Transformation at Scale

April 16, 2020 Jared Ruckle

VMware Tanzu Application Service 2.9, now generally available, is a delightful mix of something new and something familiar.

What’s new? The name, of course. Pivotal Application Service is now VMware Tanzu Application Service. The product is part of VMware Tanzu, a portfolio of products designed to help you catalyze enterprise transformation through software. (Thousands of you participated in the VMware Tanzu launch last month. Watch the event video; you’ll be glad you did!)

Along with the new name, there are dozens of new features that help you get more from the platform. The following are three capabilities worth highlighting:

  • Minimize the disruption from your chatty apps with new logging rate limits.
  • Bring more heritage apps to the platform with a custom timeout interval.
  • Move more of your .NET portfolio to Tanzu Application Service, with support for .NET Framework 4.8 and a new way to instantly modernize log streaming to the console.

What remains unchanged is the quarterly release cycle. As usual, this update is packed with handy innovations for your entire IT team.

With that in mind, read on for a review of the release highlights! (And in the meantime, be sure to sign-up for the release overview webinar scheduled to run April 23 to learn more.)

Features your application developers will love

Got chatty apps? Minimize their impact on nearby workloads with app logging rate limits (beta)

Graceful shutdowns are now even more graceful with a custom timeout interval

Java developers can deploy a sidecar with a buildpack (beta)

App Metrics 2.0: Codified custom app observability (plus Micrometer support)

Get to know Metrics Store: an enterprise time-series database that “speaks” Prometheus

Apps Manager adds new displays of metrics, routes, and sidecars

.NET developers, stream logs to the console more easily than ever before

Run .NET Framework 4.8 apps

Cloud Cache is now VMware Tanzu GemFire 

Features your platform teams will love

Streaming and historical logs are rendered more quickly, featuring an improved layout

Use Operations Manager to rotate even more certificates

Platform Automation Toolkit 4.3 cozies up to native secrets stores like CredHub and Vault

Concourse 6.0 to boost performance at scale

Got chatty apps? Minimize their impact on nearby workloads with app logging rate limits (beta)

Platform operators can now set app logging rate limits for apps running on Tanzu Application Service. Let’s explore why this is useful.

When you’re running hundreds of apps on the platform, a few of them are bound to generate an exceptionally high volume of logs. As a result, the “chatty” apps can sometimes cause logs of neighboring apps to drop. The new rate limits in version 2.9—which you configure on the “App Containers” tab—help keep those logs streaming so you can keep tabs on your apps.

Improve stability by increasing the app log limit for your noisiest apps. 

Note that you can configure different limits for isolated and/or Windows cells as well as the broader pool of cells.

Graceful shutdowns are now even more graceful with a custom timeout interval

Tanzu Application Service manages many thousands of containers in most production deployments. Automation within the platform creates and destroys containers regularly, as usage and traffic warrants. When a container does get spun down, Tanzu Application Service 2.9 now includes a new feature to make sure that container can finish processing a user request: a custom graceful shutdown period. (Previously, this timeout was set to 10 seconds.)

With this new configurable interval, which is accessed in the Advanced Features tab, even more applications will be a fit for the platform.

Configure a custom graceful shutdown period for the apps that need it.

Java developers can deploy a sidecar app with a buildpack (beta)

We’ve steadily upped our support for sidecars in the last several releases. Developers will appreciate this latest enhancement: Java apps are now fully supported with buildpacks.

Previously, developers had to roll the sidecar process into the main app, then deploy it with an app manifest. Now, the process is automated using the familiar buildpack workflow. (Note: the sidecars feature remains in beta.)

Want a quick tutorial on sidecars? Read this blog, check out docs from the community, and watch the following video.

App Metrics 2.0: Codified custom app observability (plus Micrometer support)

App Metrics, the new incarnation of PCF Metrics, displays telemetry data for the apps running on Tanzu Application Service. The module renders this data visually to help operators and developers better understand the health and performance of their apps.

Long-time users will recall the integration with Spring Boot Actuator endpoints. Version 2.0 is its most substantial update yet, a set of enhancements that improve the observability of your applications.

Cindy Sridharan, an engineer who often writes about distributed systems, noted that observability "aims to provide highly granular insights into the behavior of systems along with rich context, perfect for debugging purposes."

We’ve taken this definition to heart, adding “granular insights” and “rich context” for app developers in these specific ways:

  • Declare the metrics that matter for each app. You know best what metrics matter most for monitoring the health and performance of your workloads.  
  • Define thresholds per metric, and alert routing per app. Codify “individual warnings” and “critical thresholds” per metric. Route all app alerts to a webhook URL.
  • Share context across your organization. Indicator documents (powered by Indicator Protocol) serve as the single-point-of-truth for observability configuration, meaning dashboard customizations will persist across all users and sessions.
  • Micrometer and tagged metrics can be defined for monitoring. Micrometer delivers metrics from Spring apps to your tool of choice. App Metrics 2.0 will ingest and display Micrometer metrics effortlessly!

Create custom dashboards with App Metrics 2.0

There are easy ways to scale and update the rules that underpin the dashboards. If multiple apps have similar observability characteristics, indicator and monitor definitions can be easily augmented for multiple apps. That reduces configuration overhead for developers. Plus, application observability definitions can be version-controlled and applied via CI pipelines.

Here are three other things you’ll appreciate about these new dashboards.

  • They are timely. App Metrics can ingest and visualize metrics and logs data in near real-time, on even the largest foundations.
  • They show historical data across rolling deployments. This is a key “moment of truth” for developers. Thanks to support for CAPI v3 multi-process app architectures, you can now observe what happens during deployments.
  • Up to six weeks of data, and a smaller footprint. App Metrics can now display up to six weeks worth of logs, metrics, and event data. And we’ve reduced the storage required by quite a bit. All this data is stored in a nice and tidy new database, Metric Store.

Read on to learn more about Metrics Store, the new tile that powers App Metrics 2.0.

Get to know Metrics Store: an enterprise time-series database that “speaks” Prometheus

The brains behind the new App Metrics module? That’s Metric Store, a new component that handles all Tanzu Application Service system and app metrics.

When we announced the open-source version of Metric Store last year, here’s how we described it:

Metric Store features the auth model from Log Cache, the PromQL API from Prometheus, and the Time-Structured Merge storage engine from InfluxDB. After mixing this all together, we created Metric Store, a new component that persists all metrics from the Cloud Foundry Loggregator pipeline on its disk.

Platform teams will appreciate these three attributes of Metric Store.

  • It is multi-tenant aware. Application developers will only have access to the metrics for their apps. That makes security and compliance easier. (Operators will be able to query all metrics, as their role often demands.)
  • It is easy to query. The module is compatible with the Prometheus API and supports integrations like Grafana.
  • It has a lean storage footprint. The embedded storage engine has built-in compression to reduce the amount of storage required by a large volume of metrics. That can make Metric Store especially handy for multi-foundation deployments.

To use App Metrics 2.0, you’ll also need to install Metrics Store. Like App Metrics, Metric Store works with version 2.6 and higher.

Apps Manager adds new displays of metrics, routes, and sidecars

Apps Manager is a simple, intuitive way to keep tabs on your applications. In this release, we’ve improved the module with several useful features:

View high-level application metrics – Troubleshoot faster with quick access to essential metrics.

View app metrics with Apps Manager 2.9.

View and manage routes at the space level – Easily identify orphaned routes and diagnose app problems related to route bindings.

Manage routes with Apps Manager 2.9.

You can also now view details of sidecars attached to processes. Previously, this data was only accessible via the cf cli.

.NET developers, stream logs to the console more easily than ever before

Your apps will run better when logs are streamed to the console, rather than being statefully stored. Tanzu Application Service 2.9 makes this task much easier for .NET developers with a new tool: EventLog.Stream.Console. This NuGet package will automatically update your app to stream application and Windows event logs to the console.

Previously, you would have to update your apps manually, a process that could take up to two days to compete. Now, you can update them in minutes!

Run .NET Framework 4.8 apps

Of course you’ve already moved on to .NET Core for developing new apps, but what about the hundreds of .NET Framework applications you have in production today? Tanzu Application Service is a terrific place for them. And this release now supports .NET Framework 4.8, its final version (.NET 5 Preview 1 was recently announced).

No matter how you’re balancing your investments in .NET, Tanzu Application Service is your bridge from the old world to the new.

Cloud Cache is now VMware Tanzu GemFire

Several products earned new names. (Check out this blog for how we went about it.) Among them, Pivotal Cloud Cache is now VMware Tanzu GemFire. Why GemFire? It better reflects the product’s capabilities as a multi-threaded, distributed data and compute grid.

Tanzu GemFire is excellent for all sorts of caching scenarios—especially when performance, security, and availability matter). You can also use Tanzu GemFire as an ACID-compliant transactional data store in its own right. It can even be deployed as a serving layer for analytics!

Tanzu GemFire 1.11 shipped a few weeks back. Here are some of the things you can now do with it:

  1. Share service instances with “write” access across spaces — This is a nice data sharing feature that makes it easier for developers to update data across teams.
  2. Choose from more versions of our most popular service plans — Before you can add Tanzu GemFire to your apps, you need to add a service plan. The product offers three co-located single VM plans and three co-located multi-VM plans for you.
  3. Easily view TLS status for a service instance — Is the service instance enabled with TLS? The flag tls-enabled shows you if it is or not.

New to caching? Get started with our developer center. Got experience with caching? Then tell your story at Geode Summit at SpringOne Platform! Submit your talk proposal here.

Features your platform teams will love

Streaming and historical logs are rendered more quickly and feature an improved layout

Operations Manager 2.9 makes sifting through streaming and historical logs a little easier. Log data is now rendered faster, so you get visibility into system behavior in less time. Logs are also laid out in a more intuitive way. Your feedback was instrumental in the new design, so we’re excited for you to use this feature!

Streaming logs are improved in Operations Manager 2.9.

Use Operations Manager to rotate even more certificates

Certificate rotation is a crucial, if thankless, task. Version 2.8 made this process a little easier by rotating the certs controlled by Operations Manager for you. In Tanzu Application Service 2.9, we’ve expanded this feature to include some certs stored in CredHub as well! 

Platform Automation Toolkit 4.3 cozies up to native secrets stores, like CredHub and Vault 

Platform Automation Toolkit (formerly Platform Automation) helps you keep your platform running in a healthy state. Version 4.3, released in February, adds a new command to help in this endeavor: prepare-tasks-with-secrets. The command supports all native Concourse secrets stores, including Credhub and Vault. It has two other benefits:

Head on over to the release notes for a deeper look at the new version.

Concourse 6.0 to boost performance at scale 

In good news for Concourse community users, version 6 is available. And its top feature is better performance for users with large Concourse deployments. 

We improved the algorithm that determines which build and resource versions get included in a job.

The old process strained the database by loading each pipeline’s entire build and resource version history into memory. The new method is much better at selecting only the relevant versions. Version 6 also includes additional tweaks that improve performance for the dashboard page and overall web UI.

The new open-source version is coming soon to Concourse for VMware Tanzu, so stay tuned! In the meantime, tell us about how you’re using Concourse in this community survey.

Familiar Documentation and Upgrade Guides Help You Move to Tanzu Application Service 2.9

Already using the app platform? Then get the new version running in your sandbox environment. The documentation and upgrade guides will look very familiar!

New to the product? Browse the product page, and be sure to check out this Forrester Total Economic Impact™ study, which Pivotal commissioned from Forrester Consulting in December 2019. Into Spring? Then download A Guide to Spring Development on VMware Tanzu Application Service.

This article may contain hyperlinks to non-VMware websites that are created and maintained by third parties who are solely responsible for the content on such websites.

 Note there is no commitment or obligation that beta features will become generally available.

Cited works

 The Total Economic Impact Of VMware Tanzu Application Service, Forrester Consulting, December 2019

About the Author

Jared Ruckle

Jared works in product marketing at VMware.

Follow on Twitter Follow on Linkedin More Content by Jared Ruckle
Previous
Barbara Sanders at SpringOne Platform 2019
Barbara Sanders at SpringOne Platform 2019

SpringOne Platform 2019 - Day 2 Keynote Speakers: Barbara Sanders, The Home Depot Watch the presentations f...

Next Video
Vijay Sankaran at SpringOne Platform 2019
Vijay Sankaran at SpringOne Platform 2019

SpringOne Platform 2019 - Day 1 Keynote Speakers: Vijay Sankaran, TD Ameritrade Slides: https://www.slidesh...