-
-
Notifications
You must be signed in to change notification settings - Fork 3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
ipfs get over one hop not working #6407
Comments
If I'm understanding this correctly, the issue is that machine 1 will not fetch content from machine 3 on behalf of machine 2. This was a planned feature in the original IPFS whitepaper but has a bunch of legal implications that need to be thought through before it's practical. At the moment, machine 1 can make itself a public relay but then it'll relay traffic for any peer on the network. |
Can't machine 1 not just establish a encrypted communication between node 2 and 3 so it does not know the content it is transporting? Was already puzzled that I manually have to connect nodes to get content from A to B - if what you say is true that makes IPFS unusable for me. |
I'm assuming there's peer exchange happening on all nodes. I can't get 1 to talk to 2. They can both talk to the public peer though.
Is circuit relay operational at the moment? |
It can using the p2p-circuit ("relay") transport. However: Nodes won't relay traffic by default as this can easily run up a large bandwidth bill. However, you can enable this on node 1 by setting There's no way for node 2 to know that node 3 is connected to node 1 as well unless node 1 says so by announcing a special relay address. You can do this manually by adding a p2p-circuit address to You can also automatically use a public relay by enabling You can make it a public relay by additionally setting |
Yes (ish). Nodes will connect to other nodes through relays by default if those nodes advertise a relay. However:
|
|
to reproduce this I need 3 machines - all run go-ipfs 0.4.21:
machine 1 (lets call 'server' here) has a public IP
machine 2 (lets call 'client' here) is behind a NAT
now I can connect from machine 2 to machine 1 via
ipfs swarm connect
and get content that is pinned on machine 1now introducing machine 3 which is in the same network as machine 1 and connect them via
ipfs swarm connect
in both ways (I think there should be a way to do it one way only - see this coment: #6271 (comment)) so I can get content pinned on machine 1 by machine 3 and the other way aroundBut now when I try to get content pinned on 3 with machine 2 it is not working. It is directly downloading when I
ipfs get
the content on machine 1The text was updated successfully, but these errors were encountered: