cross-posted from: https://lemmy.zip/post/20730504
This one brings a ton of excellent updates. Take a look at the change log.
cross-posted from: https://lemmy.zip/post/20730504
This one brings a ton of excellent updates. Take a look at the change log.
Maybe open a github issue about storing the peerlist locally to fall back on if seednodes shouldn’t be reachable for whatever reason? I’m pretty sure you also can’t start a new Monero node if the seednodes aren’t available so I’d call them more of a convenience feature rather than call their existence a lack of decentralization since you can always join the network by supplying your own initial peers.
If you have a peerlist already then you do not need seednodes in monero, this makes it decentralized. If in haveno you need a seednode every time you join the network then it is not decentralized. The whole point of seednodes is to get you connected to other peers, after the the seednode is not required.
It does make sense to have a more reliable, designated entry point since you can’t be sure everyone in your peerlist hasn’t just gone offline. Also, don’t the seednodes hand out the pre-populated orderbook? As well as some other stuff? There has to be some reason why that isn’t done by regular peers, right?
The seednodes most likely perform some function that regular peers cannot due to some valid reason I am sure. This is concerning because if the seednodes are offline then the whole network goes down, thus making the network solely dependent on a server admin. The ideal is to only require seednodes for the bootstrap.
This is false. You can join and bootstrap off of any given peer that is already in the network. It will take longer to initially populate the orderbook and trade history since it comes from other peers piece by piece but otherwise you can use it completely fine.
I hope this is the case now, because I know in earlier versions when the seednodes went down and I restarted my app it would not connect even though I had previously been connected to many peers.
It was always the case but you have to manually provide the address of a peer. You could open an issue on github to save a list of known peers locally.