Managing Cloud Foundry Using Tackle

Dillon Woods on Aug 25, 2017

During a recent customer engagement, we were tasked with deploying and managing multiple Cloud Foundry environments in Google Cloud Platform. Learn how we were able to use Tackle Downstream integrated with Concourse CI to solve this difficult problem.

Pivotal Cloud Foundry (PCF) is a cloud-native platform for deploying and operating modern applications. Often companies begin their journey with PCF by picking one existing application and migrating it to the popular PaaS. Once developers have the chance to get their hands on PCF they share their experience with colleagues and adoption goes viral within the organization. This is the experience that has been repeated at Comcast, Verizon, and T-Mobile

But before developers at these large enterprises can deploy their first app, an instance of Cloud Foundry has to be installed.

Administrators who have installed Pivotal Cloud Foundry know that the installation process can be a daunting task. These are the minimum infrastructure requirements for one PCF cluster installed on AWS:

Administrators not only have to be experts on their IaaS of choice to support that much infrastructure, they must also be able to deploy and use BOSH for completing and managing the Cloud Foundry installation.

All of that is a lot to manage, but probably doable if an organization only needs one PCF cluster. The challenge really comes when a company needs multiple clusters, or even clusters that are more ephemeral in nature.

Why Tackle

Companies typically turn to Tackle when they have one of three key use cases:

  1. Solution Providers - The solution provider needs to be able to manage multiple Cloud Foundry environments, typically one for each of their customers. These clusters are often more complex than the typical environment because Solution Providers bundle additional software for their customers to use. Solution Providers may even offer managed services for these clusters in addition to deployment services.
  2. Training Environments - Once a company has completed their first application migration and has decided to adopt PCF organization wide, they find that they have a lot of new developers who need to be trained. In this scenario they may need to spin PCF clusters up and down frequently so that new developers can practice without affecting the production environment. In many cases companies choose to offer each developer their own dedicated cluster for a limited time during training.
  3. POC Workloads - SIs or ISVs often have the need to conduct Proof of Concept demonstrations using PCF. For this use case they may need to quickly deploy a cluster that mimics a production environment, demonstrate some functionality, and then spin down the cluster as quickly as possible to avoid accruing IaaS costs.

For any of these use cases Tackle can help companies in a few important ways.

Tackle can reduce the time it takes to deploy Cloud Foundry by up to 90%. It can take an engineer up to a full week to manually install and test a full Cloud Foundry cluster – Tackle can reduce this time to just 4 hours.

Tackle can reduce IaaS usage and waste. By monitoring the usage of Cloud Foundry clusters once they are installed, Tackle can help administrators track all of the environments they have running and provide alerts when a cluster is no longer being utilized.

Tackle provides self-service provisioning. Tackle makes it easy for anyone in an organization to deploy or destroy a PCF cluster. Instead of burdening the operations team, a Trainer or Sales Engineer can use Tackle to spin clusters up or down in an on-demand fashion.

Deploying Cloud Foundry with Tackle

Tackle provides a simple dashboard for deploying and managing any software product in the cloud. Anyone with access can login and provision a new instance of a product without ever having to reach out to the operations team, gain access to the IaaS account, or even understand the intricacies of installation.

In the above screenshot we can see how a user would provision a new PCF cluster in a few easy steps:

  1. The user selects the version of Cloud Foundry they want to install
  2. They choose the customer or internal business unit to associate with the cluster for cost tracking purposes
  3. A title or label is chosen for the instance (e.g. “Training instance for Brian”)
  4. Finally, the user presses the “Launch Instance” button to start the automated installation process

It’s that simple for a user to provision their own Pivotal Cloud Foundry cluster! Once the launch has been initiated users can monitor the deployment status and will be given all of the login details once the cluster is ready.

When they are done using it, decommissioning the cluster is just as easy. Pressing the “Terminate” button in the instance screen will safely destroy the instance and all associated resources:

Behind the Scenes

Tackle doesn’t try to reinvent the wheel when it comes to industry standard deployment best practices.

We fully integrate with Concourse for managing all software deployments in Tackle. This allows us to take advantage of all the features provided by a best in class CI/CD tool, such as automatic upgrades without downtime and seamless environment validation.

Since Tackle uses Concourse we can also use the official PCF Concourse pipelines provided by Pivotal for deploying PCF. Before each new deployment the latest pipeline is downloaded from Github to ensure that users are always getting the latest and greatest bits.

Tackle uses the Concourse API to push new pipelines for each PCF environment under management. In the below screenshot you can see a random name for each pipeline that indicates an individual environment.

The user never needs to login to Concourse directly, but if they do they will be able to see the status of each pipeline that has been run using Tackle. Every job in the pipeline being green indicates that the environment has been deployed successfully.

Individual jobs can be run (or rerun if needed) through the Tackle UI. A list of jobs the user is allowed to run are displayed and can be executed at any time.

Monitoring Cloud Foundry with Tackle

Tackle provides a list view that display all of the Cloud Foundry clusters that have been deployed. It includes the following information:

By clicking on an individual cluster the user can zoom in to a more detailed instance view that provides the following additional information:

All of this information provides an administrator with everything they need to monitor Cloud Foundry instances. They can see who has deployed PCF clusters, how much they are costing at an individual or aggregated level, and what they are being used for. This allows the administrator to make intelligent decisions about when environments can be destroyed.


Tackle provides a simple solution for managing multiple Cloud Foundry instances. Deployments with tackle are repeatable and reliable because of our integration with Concourse and our use of official Pivotal pipelines. By using Tackle administrators can reduce the time it takes to deploy PCF, reduce IaaS usage and waste, and provide a self-service provisioning option for their users. All environments can be centrally monitored and controlled within the organization.

Contact our sales team at to learn more!

Back to the Blog