Gathr Integrated Deployment

See the prerequisites and steps to deploy Gathr on container.

Prerequisites #

1. Verify Docker and Docker Compose #

Ensure that the latest versions of Docker and Docker Compose are installed on the Gathr deployment node.

Verify installation:

$ docker –version
$ docker compose version

2. System Requirements #

• The base OS must be RHEL 9 / OEL 9.

• Three machines are required:

  • Two for Gathr in HA (including Zookeeper, RMQ, ES, and Postgres).

  • One for HAProxy.

• Note: HAProxy and Gathr cannot be on the same machine.

• Minimum system requirements:

  • 8 CPU cores and 32 GB RAM.

• vm.max_map_count value should be 262144 in /etc/sysctl.conf.

• Python3 and pip3 are required.

• The Docker image includes Debian Linux 11 as its OS.

3. PEM File Requirement for HAProxy (Optional for HAProxy with SSL) #

• A valid .pem file is required to start HAProxy with SSL.

4. Ansible Server #

• Ansible should be installed on the server where you will run the playbook.

• Password-less SSH must be enabled from the Ansible server to the Gathr and HAProxy machines.

• Note: Password-less SSH should be configured for the root user on the remote server.

5. NFS Server #

• The NFS server must be up and running.

• Create a shared path that is accessible by both Gathr nodes.

6. Port Availability #

Ports 8090 and 9595 must be free on the machines where Gathr and HAProxy will be deployed.

Check port status:

$ netstat -anp | egrep "8090|9595"

7. User Permissions for Docker #

• Create an application user on the server with password-based authentication.

• Note: The UID and GID for this user must be the same across all machines.

• The docker and docker compose commands should be accessible by this application user.

• Ensure the application user is part of the docker group.

Verify the user’s group membership:

$ id <username>

Confirm Docker commands are working:

$ docker ps

8. Spark Cluster Requirement for Pipeline Execution #

One of the following Spark clusters is required:

• Spark Standalone v3.5.0

• Spark on YARN v3.3.3

• Spark on Kubernetes (K8s) v3.5.0

These services must be accessible from the Docker container node.

9. Local Volume for Gathr Data #

A local directory is required to store Gathr data.

Note: This path should be accessible on the Gathr nodes via NFS.

Create the directory:

$ mkdir -p /path/to/gathr-volume

10. (Optional) OpenAI API Key Requirement #

• An OpenAI API key is required to use the Gathr IQ feature.

• Internet access is mandatory for this feature.

11. (Optional) Data Intelligence (DI) #

Shared Storage for Logs and Data (Optional for Single-Node Setups) #

• Configure a shared mount (EFS/NFS) on the Docker-installed machine to store DI logs and data efficiently.

Logstash Configuration #

• Install and configure Logstash (version 6.8.23) on the chosen node.

• Ensure Logstash is integrated with Elasticsearch.

• Configure Logstash to process logs from the DI log directory.

• Logstash should run on the same node where DI logs are stored.

Access Control for Gathr User #

• Ensure Gathr user has read access to DI Docker logs.

• If access is missing, set up ACL permissions.

Load Balancer for Multi-DI Deployments (Optional) #

• Configure a private load balancer for multi-DI Docker deployments.

• Ensure it listens to the port specified in the Gathr configuration.

AWS CloudWatch Integration (For AWS Deployments Only) #

• Configure AWS CloudWatch and CloudWatch Agent for log monitoring.

12. (Optional) Mlflow Configuration #

• A private Docker registry is required.

• The Gathr team provides Mlflow images. Load these into your repository.

• K8s Cluster should be up and running.

Artifact Storage Configuration #

• MLflow generates artifacts when a model is registered, which can be stored in:

  • S3 or Ceph

  • NFS

If using S3 or Ceph, ensure you have:

• S3 Access Key

• S3 Secret Key

• S3 Endpoint URL

If using NFS, ensure you have the PVC (Persistent Volume Claim) name.

(Optional) Private Docker Registry Access #

• The machine should have private Docker registry access.

Steps to Install Gathr Using Playbook #

1. Download the Playbook Bundle #

Download the playbook bundle shared by the Gathr team on the Ansible server.

2. Extract the Bundle #

# tar -xzf GathrBundle.tar.gz

3. Navigate to the Playbook Directory #

# cd /path/to/playbook

4. (Optional) Add Host Entries Inside the Gathr Container #

Create a file named hosts in the packages folder.

Example:

# vim packages/hosts

Add the following entries:

10.0.0.1 gathr-node1
10.0.0.2 gathr-node2

Save the file.

5. (Optional) Copy HAProxy PEM File #

Copy the haproxy.pem file inside the packages folder.

This file is required to enable SSL on HAProxy.

6. Update the Properties in saxconfig_parameters #

A sample saxconfig_parameters file is provided with useful comments.

Update it with the appropriate values.

7. Reload Ansible Variables #

Run the following command to reload Ansible variables:

# ./config.sh saxconfig_parameters

8. Run the Playbook #

Execute the following command to deploy Gathr:

# ansible-playbook -i hosts gathr_one.yaml -v

Post Deployment Validation #

1. Access the Gathr UI on – https://<haproxy_hostname>:8090/

   After Gathr is up, you will see a license agreement page:

   

   Click on the “I accept” check box and click on “Accept” button.

2. An Upload license page will appear:

   

   Upload the valid Gathr license and click on “confirm”.

   

3. Click “continue” on the welcome page:

   

   Login Page will appear:

   

   You can now login with the default superuser creds:

   email - your email address password – your password

4. Gathr is deployed successfully.