This was complete news to me. So the occasional Timeout message actually means something is so broken at my end that I need to restart, even though I get Retrying messages after the Timeout one?
Do I lose my spot in the node joining line if I restart, or is the line not calculated based on the time trying to join at all?
After trying to join node overnight I woke to the following message:
Error:
0: Cannot start node (log path: /home/scott/.safe/node/local-node). If this is the first node on the network pass the local address to be used using --first
1: Could not connect to any bootstrap contact
Location:
sn/src/bin/sn_node.rs:232
Backtrace omitted.
Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.
I think this is such a cool Youtube video that I decided to upload it. I haven’t updated the Safe version on this machine, but everything went very quickly. Screw privacy. It’s a pretty old video, but I’m the guy in the checkered shirt doing unusually well.
We had an internal testnet run last night, it may be our CI / bors released a new version and checked it against a testnet. PRs are frequent and there was a PR missed in .44.
However, kinda infuriating you can’t join until you are needed (I know that’s the design and it does make sense). In real world will it work like this? Or will there be a different way to prove you are not malicious and be accepted quicker? Some sort of task?
Waiting days to join would be an adoption killer I think.
Yes, it mean the network must be able to measure that. With that it would be good to find examples.
Another consideration;/route is non Adult nodes do some measurable good network tasks, like become an archive or similar, something that needs no consensus or agreement.
However, it’s very nuanced and unclear what would work. I think it’s like bitcoin in ways, yes you join bitcoin easy, but never get paid now (probability is tiny) and here it may take a while to join but micropayments are more frequent.
So why join fast? To do good or to earn? I suppose it’s easier to come up with a scheme for the former and perhaps that is where we need to frame that research?
I would imagine the majority would want to earn as quick as possible Of course, many want to do good too.
It’s clearly not an easy problem to solve - it does seem like it needs to have some sort of solution though especially in a world where everything is instant
On another note, this network is still flying. Download seems to be getting quicker today!
Do we really want to do that? I fear for our world and instant gratification.
Perhaps we do need folk to work more? not sure, but here’s money fast feels dirty to me in some ways. However, if we can engage folk quickly then it is good to do so. Keeping in mind the whole time this is nodes and like bitcoin, these will be outnumbered by people and use of the network.
No, we shouldn’t. I more meant we should allow people the chance to work instantly, not have instant rewards. So yes, you don’t have to wait too long to join but boy do you have to work to get rewarded when you do get a chance
Yes true. I don’t know the solution - just asking some probing questions which may not be too helpful at this moment.
But instant access to do some good work where you then get rewarded for your work is what I’m getting at here
This could be of interest in relation to default timeout settings:
I previously uploaded a 600mb video, but then failed repedetly to upload a 500mb video (5 trials).
Changing timeout to 1000 (probably excessive) made the upload successful!
So it appears required timeout varies with network load or something like that.