I have an Ubuntu 16.04 instance on Amazon EC2

I can do:

geth console 2>> ~/geth.log
instance: Geth/v1.4.3-stable/linux/go1.6.1

Everything looks fine:

> eth.blockNumber

> net.peerCount

OK. Great. The blockNumber ticks up as expected.

Now, I'd like to be able to logout and go away and get on with things.

I am using screen

I do Ctrl+A followed by D. I can tail ~/geth.log and I can see the blockNumber incrementing.

I logout. However, when I tail the geth.log file several hours later, it keeps getting stuck:

I0513 05:53:21.227568 eth/downloader/downloader.go:274] Synchronisation failed: no peers to keep download active

The Amazon firewall is wide open. I have ufw running on my instance. It allows port 80, 22 and 443 (IPv4 and IPv6).

I have no idea what's going on. I have spent hours at this.

  • 2
    The geth client listens on TCP port 30303 and you'll also need UDP port 30301 for peer discovery. Try explicitly forwarding those and see if it helps. – Richard Horrocks May 13 '16 at 18:06
  • 1
    You could also monitor the connections themselves using netstat -atn (TCP) and netstat -aun (UDP), and see if/when they drop off. – Richard Horrocks May 13 '16 at 18:07
  • Thanks Richard, good suggestions. I'll try this and report back – Eamorr May 13 '16 at 22:13
  • just disable ufw and try. – niksmac May 14 '16 at 1:38
  • @niksmac Still can't get it to work. Even with suggestions from. I suspect, and I might be wrong, that all the peers I'm connecting to are in Amazon (test nodes fired up to play with Ethereum) and contain bad blocks. Hence I keep getting stuck? – Eamorr May 14 '16 at 3:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.