[Bug] No connected peer

[S3VENK]

🐛 Bug Report

Based on the provided debug logs, it appears that the issue is related to connectivity problems with a specific IP address ('MY_IP_ADDRESS'). Here’s a breakdown of the log entries:

DEBUG snarkos_node_router::heartbeat: No connected peers
    Indicates that the node/router is reporting no connected peers, suggesting a lack of network connectivity or peers available for communication.

DEBUG snarkos_node::prover: Skipping an iteration of the puzzle (no connected peers)
    Further confirms that due to no connected peers, the node is unable to proceed with certain operations or tasks.

DEBUG snarkos_node_router::handshake: Connecting to 34.17.53.129:4130...
    Shows an attempt to establish a connection with the IP address 'MY_IP_ADDRESS' on port 4130.

WARN snarkos_node_router: Unable to connect to '34.17.53.129:4130' - '3MY_IP_ADDRESS' disconnected before sending "Message::ChallengeResponse"
    Indicates a connection failure where the node attempted to connect to 'MY_IP_ADDRESS:4130' but received a disconnection before completing the expected handshake or communication step ("Message::ChallengeResponse").

Issues Identified:

No Connected Peers: The node/router is reporting no peers connected, which could indicate network issues, misconfiguration, or peers being offline/unreachable.
Connection Failure: Specifically with 'MY_IP_ADDRESS:4130', there was an attempt to connect but it failed before completing the necessary handshake process.

Potential Impact:

Operational Disruption: Tasks dependent on peer-to-peer communication or specific connections to 'MY_IP_ADDRESS:4130'  fail and delayed.
Service Availability: Services relying on these connections may experience interruptions or degraded performance until connectivity is restored.

My Environment

• snarkos v2.2.7
• rust version 1.76.0
• Ubuntu 22.04.4 LTS, CPU 16 cores, 64 GB RAM

[S3VENK]

screenshot from other day

[S3VENK]

According to the recommendations in the FAQs on Aleo's GitHub, I have already checked and ensured that ports 4130/tcp and 3030/tcp are allowed. However, the error persists.

[Demontf]

what is the network id of nowaday's testnet

[raychu86]

Same issue as #3113.

Likely an issue with the networks not being up and running or the bootstrap nodes being down. Please try again on the latest snarkOS version and feel free to reopen an issue if you experience the same problems.

[Thormeard]

The problem is still there. Running the latest available right now snarkOS version (v3.1.0). Node is in sync but then unable to connect to any peers after a restart of the process (graceful shutdown and start). After 20-60+mins node is starting to sync again and is able to catch up with the network. Was also the case for testnet-v1.1.0 and versions before that as well (although I don't remember it was being soooo slow to find any peers couple of months ago).

Steps to reproduce:

1. Get a client node that is in sync with the mainnet network
2. Restart the client
3. Client most of the times is unable to connect to any peers for a really long amount of time

Checked only for client nodes. Happens consistently for 2 different clients on 2 different linux machines. Client started with --nocdn flag.
Let me know if you are unable to reproduce the issue.

[Thormeard]

Small update, if you list peers from an already running node and then use those peers with a --peers flag, then node is starting to sync right away.

The problem is only with connecting to the 'default' hardcoded bootstrap peers. (34.105.20.52:4130, 35.231.118.193:4130, 35.204.253.77:4130, 34.87.188.140:4130).

Reproducible even on a fresh machine with no cache/no blocks. No vpn is used.