504 error on Helm installation model training

I’ve deployed Rasa X on a DigitalOcean droplet running Ubuntu 20.04 using Kubernetes and the Rasa X Helm chart.

Every time I try to train the model from the Rasa X UI (with the ‘Train model’ button), I get this error: Screenshot 2021-10-14 at 19.44.36

With the following response error:

Request URL: https://<my_domain>.com/api/projects/default/models/jobs
Request Method: POST
Status Code: 504 Gateway Time-out
Remote Address: <my_external_ip>
Referrer Policy: strict-origin-when-cross-origin

I’m not sure if this has anything to do with my problem, but the Network Policy settings on my values.yml are as follows (I enabled it because I needed to create some additional policies):

networkPolicy:
  # Enable creation of NetworkPolicy resources. When set to true, explicit ingress & egress
  # network policies will be generated for the required inter-pod connections
  enabled: true

  # Allow for traffic from a given CIDR - it's required in order to make kubelet able to run live and readiness probes
  nodeCIDR: {}
  #  - ipBlock:
  #      cidr: 0.0.0.0/0

The rasa-x-worker and rasa-x-production pods don’t seem to have any relevant logs on the issue.

Any ideas on what could be causing this error or how could I debug it?