The deployment failed because the image ghcr.io/railwayapp-templates/postgres-ssl:18-latest does not exist in the container registry. Your Postgres service is configured to pull that tag, and the registry returned "not found." You'll need to update the image source in the service's settings to a valid tag (e.g. 17-latest for PostgreSQL 17). Your existing deployment is still running normally.

ok, but I didn’t create this deployment. I followed the instructions and the guided steps from the vulnerability error, which, by the way, still exists.

To correct our earlier reply, the 18-latest tag was set by the vulnerability update flow, not by you. That tag does not exist in the container registry yet, which is why the deployment failed. Your existing Postgres deployment is still running normally. To clear the failed deployment, open the service's settings and change the image source back to the version you were running before (e.g. 17-latest if you were on PostgreSQL 17), then redeploy. The vulnerability warning persisting after the failed deploy is expected since the update was never applied, and a successful redeploy onto a valid patched tag should resolve it.

We understand you followed the guided steps, and we can see the deployment was created against the image tag postgres-ssl:18-latest, which does not exist in the container registry. The valid format is postgres-ssl:17 or postgres-ssl:18 (without the -latest suffix). Since that deployment failed, your service is still running the previous image, which is why the vulnerability warning persists. To resolve both issues, update the image source in your service's Settings to a valid tag (e.g. 17 for PostgreSQL 17), then redeploy. That will pull the patched image and clear the vulnerability notice.