`cdk deploy Supabase` fails
sagacity-felix opened this issue · 2 comments
sagacity-felix commented
Bug report
First of all, thanks for fixing the git issue in the Repository import.
When I try to deploy this stack after cloning it, it still fails due to one of two reasons.
Describe the bug
- Sometimes, the RepoImportJob exceeds the timelimit and I get:
12:46:52 PM | CREATE_FAILED | Custom::RepoImportJob | Studio/Repo/main/Default
Received response status [FAILED] from custom resource. Message returned: Error: 2023-03-30T10:46:42.596Z 502e30ef-ab43-4586-be80-dc3fb00eb348 Task timed out afte
r 180.00 seconds
Logs: /aws/lambda/Supabase-StudioRepoImportFunctionDEF0C9F1-gLGfGN4WUBfy
at invokeUserFunction (/var/task/framework.js:2:6)
at processTicksAndRejections (internal/process/task_queues.js:95:5)
at async onEvent (/var/task/framework.js:1:365)
at async Runtime.handler (/var/task/cfn-response.js:1:1543) (RequestId: 6ee1ee56-3131-4f79-b55d-b179a39a52ea)
- Other times (I think if I increase the timeout of the RepoImportJob to 5min instead of 3min), I get the following issue with the
Auth/Fargate/Serviceresource (also see screenshot):
Resource handler returned message: "Error occurred during operation 'ECS Deployment Circuit Breaker was triggered'." (RequestToken: xxx-xxx-xxx-xxx, HandlerErrorCode: GeneralServiceException)
In the ECS view in the AWS console, it just tells me that the service is unhealthy.
To Reproduce
Steps to reproduce the behavior, please provide code snippets or a repository:
- Clone the repository
- Run
cdk deploy Supabase(after signing in with the CLI) - It will successfully create almost all the resources but then fails as mentioned above
Expected behavior
The stack should be successfully created
Screenshots
System information
- OS: macOS
- Browser (if applies) [e.g. chrome, safari]
- Version of supabase-js: [e.g. 6.0.2]
- Version of Node.js: [e.g. 10.10.0]
Additional context
Add any other context about the problem here.
sagacity-felix commented
I did some more digging and it seems that the problem stems from the DB not being accessible correctly.
Here is what I get in the ECS logs for this Fargate service:
{"level":"fatal","msg":"running db migrations: Migrator: problem creating schema migrations: couldn't start a new transaction: could not create new transaction: failed to connect to `host=supabase-databaseclusterXXXXX-XXXXXX.cluster-XXXXX.eu-central-1.rds.amazonaws.com user=supabase_admin database=postgres`: failed SASL auth (FATAL: password authentication failed for user \"supabase_admin\" (SQLSTATE 28P01))","time":"2023-03-30T13:02:47Z"}
sagacity-felix commented