Node over I2P?

[btcnbegun]

do you think there are enough I2P peers for a reliable Bitcoin node? Would using I2P affect transaction propagation or block syncing speed?

[NotATether]

Obviously performing the initial block download I2P would be much slower than using clearnet, but I imagine since there is at least one I2P node ran by Core developers (it can be found in the documentation related to I2P somewhere in the Github repo of Core), it is theoretically possible to use only that peer for syncing.

Though it is definitely not reliable to only have one peer.

[apogio]

The interesting question for me is what are the advantages of running the node over I2P instead of Tor. I haven't really looked into it.

I 've read about the I2P protocol in the past and, from what I 've read:

1. The obvious difference is that on I2P you can only connect with I2P, whereas a Tor node can connect (potentially) to clearnet as well.
2. Privacy-wise I2P must be superior.

I have no idea about speed, latency, reachability etc.

[ABCbits]

do you think there are enough I2P peers for a reliable Bitcoin node?

I don't know whether there's any reliable data about total Bitcoin node that use I2P. https://bitnodes.io/nodes/ currently shows about 22.8K reachable node. But if you search ".i2p" on the search column, it shows 0 node (which should be incorrect).

Would using I2P affect transaction propagation or block syncing speed?

Yes, it's definitely slower than using clearnet. Just like Tor, I2P use multiple hops/nodes to perform communication. https://geti2p.net/en/about/performance explain about it in detail.

[apogio]

Yes, it's definitely slower than using clearnet. Just like Tor, I2P use multiple hops/nodes to perform communication. https://geti2p.net/en/about/performance explain about it in detail.

I have a general opinion that I2P isn't very actively used in many applications, am I wrong? Not just bitcoin, but generally.

But if you search ".i2p" on the search column, it shows 0 node (which should be incorrect).

I think this can be manipulated, because there are multiple occasions where I 've known about .onion nodes that weren't visible in the bitnodes website. I don't know how it's possible, but I 've seen it happening. That's why people say that the total number of nodes is actually higher than the one visible on the website.

[ABCbits]

Yes, it's definitely slower than using clearnet. Just like Tor, I2P use multiple hops/nodes to perform communication. https://geti2p.net/en/about/performance explain about it in detail.

I have a general opinion that I2P isn't very actively used in many applications, am I wrong? Not just bitcoin, but generally.

Yeah. At least, i'm sure it's less popular than Tor since I2P can't access clearnet.

But if you search ".i2p" on the search column, it shows 0 node (which should be incorrect).

I think this can be manipulated, because there are multiple occasions where I 've known about .onion nodes that weren't visible in the bitnodes website. I don't know how it's possible, but I 've seen it happening.

In this case, i wonder whether those node somehow manage to identify and ban connection bitnodes.

That's why people say that the total number of nodes is actually higher than the one visible on the website.

That's right, although usually it's stated when mentioning nodes that doesn't or unable to accept incoming connection.