Deployment strategies

Owned by Ashhad Alam
Last updated: Jun 09, 2023
3 min read

The Adeptia Connect Rancher package helps you deploy Adeptia Connect based on whether you want to use the same cluster having different namespaces for all your environments (Dev, QA, Prod), or a different cluster for each environment.

To deploy Adeptia Connect, you need to log in to the Jumpbox and execute the Adeptia Connect Rancher package by running the adeptia-connect.sh shell file (with appropriate arguments) after you have fulfilled all the prerequisites. With recommended hardware configurations, the installation of all the components and the AC application through the package typically completes in approximately twenty five minutes.

You can find the deployment related logs in installation.log file at /logs location.

The location for Adeptia Connect application error logs is /logs/debug/ which contains multiple log files, one for each microservice pods, and a pods.log file. You can open the pods.log file to identify the microservice associated with the error. Once you have this information, you can go to the log file for that microservice and see the error details.

Important!

Deploying the application requires you to have the following permissions beforehand:

  • Read/Write permission on the SSH private key file (PEM).
    To set this permission, run the following command:

    $ chmod 0600 <pem file>

  • Executable permission on the adeptia-connect.sh shell file.
    To set this permission, run the following command:

    $ chmod +x adeptia-connect.sh

Deploy to multiple clusters

The following diagram is a graphical representation of AC deployment on multiple clusters.

Run the following command to deploy the application and all the other components to a cluster. Use the same command to deploy the application (to be used as another environment, for example, QA) to another cluster.

$ sudo ./adeptia-connect.sh

Important!

If you are using encrypted vault-config.yaml file, you need to pass the argument --ask-vault-pass while executing the shell file as shown in the example below:

$ sudo ./adeptia-connect.sh --ask-vault-pass

Deploy to multiple namespaces in a cluster

The following diagram is a graphical representation of AC deployment to multiple namespaces within a cluster.

If you want to keep all your application environments (Dev, QA, Prod) in a single cluster, you need to deploy the application to different namespaces within the cluster.

To achieve this, you first need to deploy all the components except for Adeptia Connect, and then deploy the application individually to each namespace.

Follow the instructions given below:

  1. In the general-config.yaml file, do the followings:

    1. Set the namespace to which you want to perform the install operation.

    2. Set the domain for accessing the Rancher UI.

    3. Set the domain for accessing the application.

  2. Run the following command.

    This command installs all the components except for Adeptia Connect to the cluster.

  3. Run the following command to install Adeptia Connect.

Using the tag argument in the install command

You can use the tag argument while running the command to execute the shell file for installing different components as per your requirement. For example, if you want to install all the components except for the Adeptia Connect application, run the following command.

Also, you can use multiple tags separated with commas as shown below:

The following table contains the the list of some tags and their description:

Tag

Description

Tag

Description

--tag=install-all

Installs all the components in one go.

--tag=install-basic

Installs all the components except for the Adeptia Connect application.

--tag=install-ac

Installs Adeptia Connect application only.

Introducing additional nodes to the cluster

In a cluster, pods share the resources available across the nodes. In case the available resources are not enough for the application to run smoothly, you may want to introduce additional nodes to the cluster. You can run the following command to introduce a node to the existing cluster.

Before you run this command, ensure that:

  • RKE2 server is running.

  • The domain name or IP address of the VM/RKE2 agent you want to introduce is added to the inventory file.

 

 

 

Â