Remove DNS Seeds run by entities which were never well-established. #4671

As there is a clear policy DNS Seeds should be adhering to, they also need to be run by community members which are semi-trusted to act according to this policy. These two nodes were submitted without any clear evidence of community trust in their organizations and, as such, should either provide such evidence, or be removed. @ayeowch and @bitkevin were the original submitters of the pull requests which added these nodes. Note that they are both part of projects to add more full nodes to the network, which they can continue to do with or without a DNS Seed.

Also, blog.open-nodes.org seems to indicate they are running nodes with aggressive outbound connection counts, amounting to attacking the network and replacing diversely-operated nodes with ones they run :(.

Addition of seed.bitnodes.io was discussed and ACK'ed in #3875. This is part of getaddr.bitnodes.io infrastructure sponsored by the Bitcoin Foundation and operated independently by myself. I believe most of the seeders are operated based on sipa's bitcoin seeder (https://github.com/sipa/bitcoin-seeder). seed.bitnodes.io is based on a separate Python implementation available from https://github.com/ayeowch/bitnodes/blob/master/seeder.py. Nodes with most common height are selected on random basis and served via NSD authoritative-only DNS server. The seeder is currently hosted on a dedicated server by Hetzner in Germany and has been one of the most healthy according to http://bitcoinstats.com/network/dns-servers/. I'd be happy to provide more evidence if so required in order to continue the operation of the seeder.

I'm really only concerned with who the people are who are running the infrastructure. They need to be established within the community. ie before bitnodes and open-nodes they were run by developers who had been working on Bitcoin Core and had established trust that they were both technically competent and were acting for the good of Bitcoin. Though I dont propose that only developers can run a DNS seed, running a DNS seed is not something that someone who has no previous experience in the Bitcoin development community can do.

Of course I appreciate everything people are trying to do, and though running a seed is useful, its not the most critical thing for the network, especially after #4559.

I don't think Bitcoin Core needs to care overly much about this given that it has relatively little dependence on the seeds. SPV clients rely on the seeds more, but I'm personally satisfied by Addy's contributions at least. The other one (opennodes) I haven't seen and isn't in bitcoinj: I have no opinion on that one. But my gut sense is: more is better.

"must be run by entities which have some minimum level of trust within the Bitcoin community" I'm not sure how to judge this. But I agree with @mikehearn here, it seems bitnodes.io gets closer to it than open-nodes.org.

So - let's move this along and remove open-nodes.org.

Fair enough, removed only open-nodes.org as bitnodes is only not well established to those who weren't paying attention while they were getting established.

ACK

Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/p4671_550d4fa7a77c9763d4de9e0c0c48bc2655a65017/ for binaries and test log. This test script verifies pulls every time they are updated. It, however, dies sometimes and fails to test properly. If you are waiting on a test, please check timestamps to verify that the test.log is moving at http://jenkins.bluematt.me/pull-tester/current/ Contact BlueMatt on freenode if something looks broken.