Topic: The fall of Bitcoin nodes. Why nobody talks about it ? (Read 6916 times)

Price is one thing. Availability and service is another thing.

For example, if your node goes down on a friday evening, will there be staff to get it back up before monday morning? Perhaps you even run other stuff on that VPS/dedi, like a homepage or other services.

Geographical location is another issue. In the US, prices are low, because of very high competition between datacenters and hosting companies. In addition, it must be assumed that the NSA eavesdrop on all communications on US soil, if you bought a VPS with your cc, the NSA knows your home ip (they derive it from the connections you make to that VPS), they know the ip of your node, and they also most likely know all the other purchases you've done with that cc. Ie. you're operating in bright daylight in sight of everyone.

If for whatever reason, there was a crackdown in the US on bitcoin, it would not be unlikely that all dedi and VPS'es running a bitcoin node in the US would be shut down. In addition your cc might be blocked, and you might get a visit of a 3-letter agency. I'm not saying this is going to happen, I'm just saying it might happen.

In such a situation, it's important that there's plenty of nodes outside the US. For that very reason, it's good for node-operators to operate nodes outside the US. To aid in more decentralization and robustness for the network.

As for Linode, please do not forget the bitcoin businesses that were the victims of their poor security practices. infamous Linode hack

Security is our number one priority and has been for over eight years. We depend on and value the trust our customers have placed in us.

Yet, one of their web-based systems was compromised, with the result of a large stash of bitcoins vanishing.

Relevant read: https://bitcointalksearch.org/topic/response-from-linode-regarding-the-theft-case-73640

One could however argue that it was stupid to keep bitcoins on those servers in the first place, and I agree - but the treatment from Linode's side was less than stellar imo.

I am working on a project to raise the node count.

No problem.
It's true that Linode used to be overpriced but since about 1 year they matched DigitalOcean performance.
Amazon AWS is still a little bit more powerful for the price but I can't stand their interface.

The number of full nodes will go up and down throughout the day (and week, and month) so a change like that in a short period would not worry me.

The fact that many nodes are located in people's homes means that many does will be turned off at various times for a number of reasons (going to sleep, going to work, going on vacation, power going out, internet interruption, needing to use computer for some kind of data/CPU intensive process) only to be turned back on once the reason it was turned off is resolved.

The distributed nature of the network means that when one node goes down/off, the other nodes will take their place

What I was saying is that for $40 you can get a much better cheap dedicated than what vps providers offer. I just checked and Linode is now cheaper than it used to be, it seems competition from Digitalocean made them rethink their plans and pricing. So forget what I said about linode being "way overpriced"

I'm not sure if the bitnodes script sees all nodes either.

But I've never seen anyone address how many nodes should the network really have to be big and strong? Do we need 10000? 100000? More geo-distribution?

Since your post down to 6500.. 

I have seen people offering to run a node in the name of someone who pays them as little as .04 per year - they would essentially be doing all the 'legwork' to running the node, but the buyer would essentially be covering all the financial costs

I just use Linode and DigitalOcean. What are the way cheaper and better VPS that you talk about ?

I remember one really cheap VPS, I ordered one to try, 2 months after the company has gone bankrupt and I lost my money.

Yeah, you can get a nice dedicated for $40 these days... Linode is way overpriced.

Hm.. I've toyed with the tought of mounting disk space over the network on VPS'es with too little disk-space. Would that be feasible?

no one can make a change to it

The problem are the 30GB required and more to come. If you have a 40GB HDD for $15 per year, please share it because we never heard about it.
Like I said, I use my VPS to mine Monero. For a long time I was making a profit from it, now it helps me reduce the bill.

$40 a month is a lot to spend on a full node. There are much cheaper options out there (some as low as $15 per year). As of now, you really do not need a very high performance computer in order to run a full node so there is no real reason to get such a high performance VPS

Always 7000 nodes here ... no problem : https://getaddr.bitnodes.io/

Is it just me or are most people on this post missing the point?

The problem is not whether we can host more nodes or not. The problem is that it is not being done, even if it is easy. Why? because there's no incentive for most people, in fact there's negative incentives to do it. Aka blockchain bloating and no immediate benefit.

The idea that saying things like "people ought to support the network our of a none financial sense of reciprocity" is sadly naive, even if fair.

Mining is also more centralized than ever and this is becoming worse over time with not expectation of reversing. Right? Since all the rewards go to miners, node hosting becomes charity.

How do we solve this?

Well, the first question is, does it matter?  What are we losing with the centralization of nodes? is it a worse problem than the centralization of mining?

if we decide more nodes is better than less, than one option is to say, something like 10% of mining rewards per block go to a random node, or are distributed among x random nodes.

Sort of like Darkcoin's Master Nodes are financially rewarded.

Would this centralize mining more or make some miners go out of business? I don't know. I frankly think Mining is unnecessary, Proof of Stake and similar innovations have already solved this problem, or atleast made it less of a problem.  particularly with appropriate early distribution of funds.

Ha. good point 

I guess I was trying to say that my node will be running for a long time.  The bandwidth isn't a big deal.  Right not it is about 100 Gb/month upload and ~30 Gb harddrive space.  I'm running my on a older computer that powered partially by solar during the day so actual power draw is minimal.

Sorry, I was looking for the cheapest and I didn't look at the HDD.
I personally use this one :



It is damn super fast and mining Monero with it greatly reduce the bill.

Technically, as long as your node is up bitcoin can't be considered dead .

I will be running my node until the day that Bitcoin dies.