Non-validator Node Timeout Deadline Too Fast

[BigBenlau]

I have deployed a non-validator node in Tokyo as per the recommended setup. However, the node fails to start syncing due to a timeout deadline that appears to be too short. The following error messages are observed in the logs:

2025-06-11T14:11:04.621Z WARN >>> hl-node @@ reading bytes for abci_stream greeting: 680000000/757521515

2025-06-11T14:11:04.939Z WARN >>> hl-node @@ reading bytes for abci_stream greeting: 684000000/757521515

2025-06-11T14:11:05.257Z WARN >>> hl-node @@ reading bytes for abci_stream greeting: 688000000/757521515

2025-06-11T14:11:05.575Z WARN >>> hl-node @@ reading bytes for abci_stream greeting: 692000000/757521515

2025-06-11T14:11:05.609Z WARN >>> hl-node @@ abci_stream from 199.254.199.12 timed out: lu::timeout deadline has elapsed

2025-06-11T14:11:05.609Z WARN >>> hl-node @@ error connecting to candidate peer Ip(199.254.199.12) with greeting TcpGreeting { send_abci: true, send_evm_kvs: false, id: Live }: Timed out connecting to peer

Is there a way to adjust the timeout duration to allow the node to successfully connect and start syncing? Any guidance on configuration changes or potential workarounds would be appreciated.

[sgrsaga]

I also face the same issues. Have tried many peers but it failed to come up. After few hours it comes back up again.

Sample error messages:

hl-visor[7339]: 2025-07-11T11:09:00.396Z WARN >>> hl-node @@ reading bytes for abci_stream recv greeting: 560000000/983193498
hl-visor[7339]: 2025-07-11T11:09:00.632Z WARN >>> hl-node @@ reading bytes for abci_stream recv greeting: 564000000/983193498
hl-visor[7339]: 2025-07-11T11:09:00.874Z WARN >>> hl-node @@ reading bytes for abci_stream recv greeting: 568000000/983193498
hl-visor[7339]: 2025-07-11T11:09:01.123Z WARN >>> hl-node @@ reading bytes for abci_stream recv greeting: 572000000/983193498
hl-visor[7339]: 2025-07-11T11:09:01.358Z WARN >>> hl-node @@ reading bytes for abci_stream recv greeting: 576000000/983193498
hl-visor[7339]: 2025-07-11T11:09:01.603Z WARN >>> hl-node @@ reading bytes for abci_stream recv greeting: 580000000/983193498
hl-visor[7339]: 2025-07-11T11:09:01.690Z WARN >>> hl-node @@ abci_stream from 18.181.113.192 timed out: lu::timeout deadline has elapsed
hl-visor[7339]: 2025-07-11T11:09:01.690Z WARN >>> hl-node @@ error connecting to candidate peer Ip(18.181.113.192) with greeting TcpGreeting { send_abci: true, send_evm_kvs: false, id: Live }: Timed out connecting to peer
At least need a method to identify healthy peers

If I'm correct this is failing because we fail to the initial handshake with the peer which is about 983193498 byte (over 900 MB) and peer time out.
Why such a huge handshake stream? Is there anything we can do to stabilize this ?

[franceschiniandrea]

any update on this? How can we increase the timeout deadline when connecting?