Is that in reference to rewards? If so, I’m not thinking rewards, I’m thinking gates to filter attacks. Otherwise again, why would an attacker not flood the join queue? (or queues if they’re sectional)
I now have a Safe Network node running in the cloud
, waiting for chunks to store ![]()
Next is to try and set-up SBCs (Odroid and Raspberry Pi) but I’m tired after a busy day so it will have to wait!
You checked the price of scrap lead these days? Don’t give them away at the skip. Take them to the scrappy and get yourself some decent* beer money.
*If such a thing as decent beer exists round your parts. Now a nice sparkling glass of cider… They do that VERY well indeed.
Concerning hangs and random ports…
Sadly it looks like constant port just make hang more probable.
This time I am running random ports and I still got not-joined and not-refused situation: sn_node.log (DropMeFiles).
No excuse whatsoever — have a nice glass of sparkiling zoider and get to it!!!
I’m for a wee try at that - thank you ![]()
but I’ll reduce the sleep 30 to something vicious like 2
Looks like it is possible to flood network with problematic nodes in any case ![]()
Do you mean paying to be able to farm? Smells like scam.
Also I do not think that it can stop TLA / Google / etc.
How does someone earn in the first instance though?
That’s not easily possible as, with any competent ISP, your neighbor would perform better than colocated equipment. Geographic distribution is what Big Tech wants (edge CDN et al) but the people possess.
What about botnets?
I’m describing attack scenarios given my understanding of the designs. You’d have to be more specific for me to be able to be.
Skin in the game, is one solution… that you do have to buy option to run a node. It’s not ideal for those without but would be a safe method for startup. Those without coin to start with, can have more appetite for joining after stability is achieved. The cost is offset after a time but would be an inhibitor for those trying to pwn too many nodes.
How does someone earn in the first instance though?
Obviously on an exchange, but also maybe an invite system - send some tokens to friends. Exclusive and slow start, but IMO would grow alright if the tech is working well.
What if every queue selection for node hosting required burning e.g. 10 safe to the network as an anti-spam measure?
That’s an interesting idea. It wouldn’t be required for anonymous read use of the network, which should always be no cost, so it’s viable.
Perhaps it could be extended beyond enqueuing. There’s a known issue where a node with age could lose their seniority if they have a power outage or are moving and will be offline for several days. Some SNT could be put up as a bond, holding the node’s spot while offline. It’d be a way of incentivizing planned maintenance windows for network nodes, rather than having nodes just disappear.
send some tokens to friends
There’s probably value in nodes largely funneling into the network through the social network of Safe supporters.
I have documented the steps I made on Ubuntu 18.04 here under Section 10, and I am stuck, not being able to create a Safe. Executing the steps under [FLEMING_INSTALL_LINUX 20210409], and [SETUP_AUTH_AND_SAFE 20210409] leads to “Error: AuthdClientError: [Error] ClientError - Response not received: read error: connection closed: timed out”. Time for a break, and try later again…
Obviously on an exchange, but also maybe an invite system - send some tokens to friends.
To me, that breaks Fundamental number one though: requiring a 3rd party to use the Network.
Whether it’s that or anything else though, perhaps there is no harm in having slightly compromised fundamentals initially if it helps the network get established.
Tough sell. It’s still just concentrating ownership of nodes to those with more access to money too.
I’m sure we’ll find a better way that won’t require compromising either of those things.
I’m curious what weakness you or others see in:
As for useful resource_proof ideas, would checksumming a random data/block qualify? I’m think of it like a ‘free,’ externally performed and validated, continuous ZFS scrub. Using the scenario of a node joining the queue, the node is given some URL and a hash which can either be what’s expected or a rand() fake, the node performs a read of the URL, calculates the hash of its data, returns whether the hashes match, and if the answer is correct the node is entered to the queue.