This message was deleted.

Fixing HTTP 504 Error When Saving MySQL Source in Airbyte 2.0.18 on Azure AKS The 504 Gateway Timeout error when trying to save/test a MySQL source is a known issue in Kubernetes deployments. It typically occurs because the ingress timeout is too short (default 60 seconds) for the source check/discovery operation to complete. Here are the most likely causes and fixes: 1. Increase NGINX Ingress Timeout (Most Common Fix) The default NGINX ingress timeout is 60 seconds. Add timeout annotations to your ingress resource:

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: airbyte-ingress
  namespace: airbyte
  annotations:
    nginx.ingress.kubernetes.io/proxy-read-timeout: '600'
    nginx.ingress.kubernetes.io/proxy-send-timeout: '600'
    nginx.ingress.kubernetes.io/proxy-connect-timeout: '600'
spec:
  ingressClassName: ingress-nginx-internal
  # ... rest of your spec

This is explicitly recommended in Airbyte's documentation: "Set appropriate backend timeout values for the Airbyte server ingress. Timeout values that are too short can lead to 504 errors in the UI when creating new sources or destinations." [Ingress Docs] 2. Check for Missing Workload API Auth Token (Known Bug in 2.0.1/2.0.19) There is a known issue where

connector-sidecar

pods don't receive

WORKLOAD_API_AUTH_HEADER

and

WORKLOAD_API_BEARER_TOKEN

environment variables, causing 401 errors that surface as 504 timeouts in the UI. [GitHub Issue] Check if your connector-sidecar pods have the auth tokens:

kubectl get pod <source-mysql-check-pod-name> -n <airbyte-namespace> -o json | \
  jq -r '.spec.containers[] | select(.name=="connector-sidecar") | .env[] | select(.name | contains("WORKLOAD")) | "\(.name): \(.value)"'

If

WORKLOAD_API_AUTH_HEADER

and

WORKLOAD_API_BEARER_TOKEN

are missing, ensure they are set in your `values.yaml`:

workloadApiServer:
  extraEnv:
    - name: WORKLOAD_API_AUTH_HEADER
      value: "Authorization"
    - name: WORKLOAD_API_BEARER_TOKEN
      value: "<your-token>"

worker:
  extraEnv:
    - name: WORKLOAD_API_AUTH_HEADER
      value: "Authorization"
    - name: WORKLOAD_API_BEARER_TOKEN
      value: "<your-token>"

workloadLauncher:
  extraEnv:
    - name: WORKLOAD_API_AUTH_HEADER
      value: "Authorization"
    - name: WORKLOAD_API_BEARER_TOKEN
      value: "<your-token>"

Also ensure

API_AUTHORIZATION_ENABLED=true

is set in your ConfigMap. [GitHub Issue] 3. Check for Pod Creation Failures Verify there are no Kubernetes-level issues preventing workload pods from starting:

kubectl get events --namespace <airbyte-namespace> --field-selector reason=FailedCreate

kubectl describe nodes | grep -A 5 "Allocated resources"

kubectl logs -l app.kubernetes.io/name=workload-launcher --namespace <airbyte-namespace> --tail=100

[GitHub Issue] Summary Start with the ingress timeout fix as it is the most common cause of this specific error. 👍 0 button 👎 0 button ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help., with interactive elements