Devel sync-mainnet stuck on startup (can't find peers)

[uberspot]

Hi,

i'm having an issue with running the example from the devel/sync-mainnet/ folder and all links to your discord seem to be expired in docs so I can't find/join it.

Lighthouse release used: https://github.com/sigp/lighthouse/releases/tag/v3.3.0
Config used(no changes, just for testing): https://github.com/streamingfast/firehose-ethereum/tree/develop/devel/sync-mainnet

streamingfast/go-ethereum release used: release/geth-v1.10.26-fh2.1

fireeth --version
fireeth version dev

Issue: Basically the geth node can't find peers. It gets stuck in this state for >30minutes and then I terminate it. I checked the security groups in AWS and they all seem reasonably open for the ports it needs. I tried opening the ports completely too to 0.0.0.0/0. Other aws compute with rocketpool and erigon nodes running in the same security group work fine so I don't think it's the network.

For testing:

export CHECKPOINT_SYNC_URL=https://sync.invis.tools

Consensys output: https://pastecode.io/s/djzjdw8i

Firehose output: https://pastecode.io/s/ym9pfdkk

[maoueh]

From the consensus output, it seems that it was never able to contact the execution client (geth here managed by firehose).

All I see there is Connection refused errors. Are you able locally from the machine that is running lighthouse to contact the execution endpoint? Maybe they are running on two different machine and hence, you would need to update http://localhost:9551 in the lighthouse config?

[maoueh]

Nothing we can do, peering issue is something outside the scope we can deal within Firehose. https://geth.ethereum.org/docs/fundamentals/peer-to-peer might help, I've added that to the doc.