@kapa.ai I am deploying airbyte 2.0.7 using helm command . When i deploy it, the helm command fail on pre-upgrade hook with the bootloader pod failing. The bootloader pod is failing because it is timeing out ERROR i.m.m.h.i.HealthResult$Builder(exception):123 - Health indicator [airbyte-db-svc:5432/db-airbyte] reported exception: java.sql.SQLTransientConnectionException: HikariPool-1 - Connection is not available, request timed out after 30000ms (total=0, active=0, idle=0, waiting=0) java.sql.SQLTransientConnectionException: HikariPool-1 - Connection is not available, request timed out after 30000ms (total=0, active=0, idle=0, waiting=0) This is possibly because the airbyte-db-svc is not running in the namespace. Shouldnt the service be initialised on its own as part of helm steps or am i missing something here ?

Based on the knowledge sources, the error you're encountering is a common issue during Airbyte deployment via Helm. The bootloader is failing because it cannot connect to the

airbyte-db

service. ## Root Cause The issue is typically that the

airbyte-db

pod is not ready or running when the bootloader tries to connect to it. According to the [troubleshooting documentation](https://docs.airbyte.com/platform/deploying-airbyte/troubleshoot-deploy), "Failures reasons are generally related to problem of

airbyte-bootloader

not able to connect to the

airbyte-db

service." ## Common Scenarios From the knowledge sources, several scenarios can cause this: 1. Database pod not starting: The

airbyte-db

pod may be in a

Pending

state due to volume provisioning issues. In [one case](https://discuss.airbyte.io/t/3127), the database pods were pending because persistent volumes were waiting to be created by the external provisioner. 2. Database initialization issues: The database container may be failing to start due to permission issues or initialization errors, as seen in [this issue](https://github.com/airbytehq/airbyte/issues/54199) where PostgreSQL had invalid directory permissions. 3. Connection refused: The database service may not be accepting connections yet, as shown in [this GitHub issue](https://github.com/airbytehq/airbyte/issues/68136) with

Connection refused

errors. ## Troubleshooting Steps To diagnose your specific issue, check: 1. Pod status: Run

kubectl get pods -n <your-namespace>

to see if the

airbyte-db

pod is running 2. Pod events: Run

kubectl describe pod <airbyte-db-pod-name> -n <your-namespace>

to check for errors 3. Database logs: Run

kubectl logs <airbyte-db-pod-name> -n <your-namespace>

to see if there are initialization errors The

airbyte-db-svc

service should indeed be created automatically as part of the Helm deployment. If the service exists but the pod isn't running, that's where your issue lies. The bootloader is designed to wait for the database to become available, but it will eventually timeout (as you're experiencing) if the database never starts successfully.

yup i dont see the pods for db in the namespace - what would be the fix then ? @kapa.ai

A message from kapa.ai