Morgan Davies

Kong now offers Mesosphere DC/OS. This integration enables developers to deploy Kong on a Mesosphere DC/OS cluster to simplify operations and achieve higher resource utilization.

Apache Mesos is the open-source distributed systems kernel at the heart of DC/OS. By abstracting the entire datacenter into a single pool of computing resources it simplifies running distributed systems at scale. Mesosphere can support different types of distributed workloads and container orchestration, like Mesos and Docker containers, and stateful big-data technologies, such as Cassandra, one of Kong’s optimized DB options.

Kong is designed to sit in front of highly performant APIs and microservices to protect, extend and orchestrate distributed systems. With nearly 12,000 stars on Github and 4 million plus downloads, Kong is the most popular open-source API gateway and microservices management layer.

Kong + Mesosphere Meetup – October 25th. Register here to join the event!

“Kong always welcomes ecosystem partners and is especially excited to work with Mesosphere. Container orchestration, with DC/OS, powers modern applications with microservices and data services; a perfect fit with Kong. With this integration, developers can now provision Kong on a DC/OS cluster” said Marco Palladino, Kong’s CTO.

The integration with Mesosphere will enable Kong to offer an official DC/OS package. Learn more about Kong deployment on DC/OS here, or follow the simple GUI based steps outlined below.

Kong can be provisioned on a DC/OS cluster using the following steps:

1. Deploy a DC/OS Cluster

Skip to step 2 if you have already provisioned a DC/OS cluster.

Following the DC/OS AWS documentation deploy a DC/OS cluster where Kong will be provisioned. Once the cluster is ready, Kong can be deployed using the DC/OS CLI or the DC/OS GUI.

2. Deploy Marathon-LB

Select Marathon-LB for load balancing external traffic to a cluster. Use the package marathon-lb to deploy the Marathon-LB:

3. Deploy a Kong-supported Database

Before deploying Kong, provision a Cassandra cluster or PostgreSQL instance.

For Cassandra use the `cassandra` package to deploy 3 nodes of Cassandra in the DC/OS cluster. Below is an example of provisioning a PosgreSQL instance using the `posgresql` package from Universe.

Devs can update the configuration properties in `Advance Installation`. Please note the following requirements while configuring the Kong package: Hostname, Username, Password, Database name.  

4. Deploy Kong

With an external load balancer and Kong supported datastore running, simply deploy Kong using the `kong` package from Universe repo.

Based on the database and network preferences, update the configuration properties in ‘Advance Installation.’


5. Verify your Deployments

From the DC/OS UI there should be 3 services in ‘running’ state.

6. Using Kong

Now that Kong is installed, test the configuration SSH into one of the instances in the cluster (such as a master) and try curling the endpoints:

$ curl -i -X GET http://marathon-lb.marathon.mesos:10201

Quickly learn how to use Kong with the our 5-minute Quick Start guide.


Share this: