blockscout-frontend in ethereum-package is not reachable while port forwarding from a k8s cluster or VM to localhost

[TobiWo]

Hi there,

I'm running a local ethereum devnet using the ethereum-package for kurtosis. Sometimes I want the network to last longer which is where I deploy it in K8s or on a VM (via docker). Now, if I port forward the blockscout-frontend port to my localhost the frontend is not reachable. However, it works when I simply run the devnet on my local machine. I mean I do not need to port forward in this case and then blockscout is reachable. I thought that it might be the env variable NEXT_PUBLIC_APP_HOST which is set to 127.0.0.1 in the ethereum-package but even when I set it to 0.0.0.0 I still have the same issue. I easily can port forward the CL http port or the EL rpc port or even the DORA explorer port which is why I think this is not ethereum-package related. I'm a bit baffled. Any help is much appreciated.

Best,
Tobi

[ekawibisono]

hello @TobiWo , You build use kubernetes ? or docker-compose ?

[TobiWo]

We use K8s but as mentioned it is not a separate installation but installed via the kurtosis ethereum-package. However, I might try a separate installation via the helm chart.

[TobiWo]

Making some progress here. I managed to access the frontend while setting HOSTNAME env in the ethereum-package config. However, it looks like that the indexer is not running correctly as well. The database is empty and the container is producing these logs:

{"time":"2025-04-16T15:01:37.611Z","severity":"info","message":"Index had to catch up.","metadata":{"fetcher":"block_catchup","last_block_number":0,"shrunk":false,"missing_block_count":1,"first_block_number":0}}
{"time":"2025-04-16T15:01:37.611Z","severity":"info","message":"Checking if index needs to catch up in 0ms.","metadata":{"fetcher":"block_catchup"}}

Will check if this is also the case without the HOSTNAME env var but I guess it has nothing to do with the indexer issue.