Description
Ask a question
When we used the recommended command from the README to start the entire stack and deploy the GraphQL service on the preprod network, we noticed that the background container kept displaying "metadata server unreachable" and frequently restarting. This issue directly prevented us from retrieving data through the GraphQL service.
So, we tried explicitly specifying the following during the restart:
METADATA_SERVER_URI="http://token-metadata-registry:8080"
TOKEN_REGISTRY_PORT=8080
After the restart, we found that the background container ran normally, and after three days of observation, we did not encounter any similar error.
We noticed that in the docker-compose.yml file, the default values for METADATA_SERVER_URI and TOKEN_REGISTRY_PORT in the background container were exactly the same as what we had set (in fact, we are using the default values).
I would like to know if anyone else has been able to reproduce this error? Perhaps we should have explicitly specified these values instead of relying on the default ones shown in the docker-compose.yml file?