Depends on exactly how you ran the original command but most of the time I would expect closing the terminal to close any processes associated with that terminal.
Yeah, if you’re running the node directly as a foreground process it would die along with the terminal.
The launch tool will start the nodes as background processes.
I would say at some point we should put together some kind of setup/install/package that sets up and runs the node as a systemd service or something along those lines.
Not as yet. But we’re working on it.
Also genesis key is indeed needed. This will ship as part of the network config (along w/ node ips). This lets clients/nodes verify their on the correct network and being handed valid data.
@chriso the log info and suggestions, heck just the format in general is looking super user friendly and easy on the eyes.
I did want to point out that
SafeCoin is still in there though and I thought there was a push to move to
Safe Network Tokens for quite a few reasons. Not wanting to nitpick and I’m not privy to insider talk so just relaying what I can follow along with in any updates.
Amazing changes all around!
Edit: props @chris.connelly!
I suspect that farming is broken at this moment.
I am wondering if this is going to be the case for a live network when people want to install a node to join?
How is the regular joe going to know if the key is for the global live network or some scammer who they got the node software off. (like a rouge site giving it as a download or a other way). The key being a string of numbers/letters is not human friendly. The early testnets had a string of text to identify the network.
I realise this has security benefits and so on. But the human readable seems important for the human to know his/her node is on the network they want.
Will there be a way to verify which network the genesis key connects to in a way that cannot be scammed. Like going to MaidSafe.com and verifying the key. The network self identifying is scammable I would think since the fake one calls itself the real one.
Maybe the genesis key is generated from a text string that the human can easily do in order to verify.
Could be yeh. UX of it hasn’t been thought through at all really yet. One for @JimCollinson down the line perhaps.
You should no more need to see the key than the ips to nodes tbh, its all part of config. The key is not inherently more untrustworthy than the IPs. And with that you have to trust the source of that file (and that that has not been tampered with). So atm, I guess it’s similarly usable to hashing files for verification of integrity, with all that’s usability failings.
I think the ability to be able to generate this file and hand it to friends may well be useful. Therefore not trusting a site (and handing off your latest network knowledge to them too). At least they’d be in the same network as you, whether that’s MAID sanctioned or no…
Yeh, I was thinking of being able to see the network you’re connecting to as a visual conformation.
Once the network is live I am sure there will be the scam emails and so forth promoting a network that is not Safe but another for whatever purpose the person has. Who knows what for.
Just nice to see the text and know its the right one.
Maybe distribute the key as a text and then its hashed to use in the node while running. So when the startup code is reading the config the text is hashed.
Basically the less knowledgable person is not going to know what IPs to trust and the text is a good confirmation of the network they are using. Could be a corporate network and need to be sure not connecting to the global network LOL
Not an issue with the test nets or early network when its knowledgable people using it. And yes I am thinking ahead.
Thanks for your reply.
If user do not know which IPs to trust, he will not know to what text he should trust too.
Why? He learns of the safe network with its named network. Gets software and the client tells him the network name.
Yes you can get deeper and say the scammer will modify the software, but at that point nothing will protect them. At that point not finding their friend’s uploaded file will be a good indicator.
Yes there are obvious levels of scamming, but the easiest is to simply change the network key & IPs.
Also the case where there are specialised networks for whatever reason, having a readable text name for the network makes it easy for the user to realise they have the wrong network when the UI shows “XYZ Company network” rather than “Safe Global Network” OR “testnet release candidate 1”
Yes, user either get trusted binaries & config or non-trusted.
If he chooses non-trusted version, then anything may happen.
(Trojan, which will steal tokens, for example)
Thanks! That’s courtesy of the color-eyre crate. We’re using that to improve the error handling and user experience.
Chores done, where’s my dev update?!
No dev update until I polish the agogometer.
Polished, calibrated and deployed.
I looked at
and the needle immediately rose up like a thing that rises up to reach a throbbing 8.73.
I’M never going to get first with you lot hanging about
Stick at it, I’ve been looking on the from of my boat since posting.
What is a spot that needs to be differentiated from a blob
Ahh I see now.
Blob is at least 3072 bytes, as that is min self encryptable bytes. A
Spot is smaller than that, and will be encrypted with the client encryption keys. Both of these are type safe in that the construction of them will validate their size.”
anybody know if defi websites would be feasible in the safe network? that I think would be great. because ATM defi websites are still more or less able to be tracked and shut down via their finding the host servers.
Cool stuff on denominations if you stalk the GitHub. Not posting anything so I don’t spoil anything that might be part of the Thursday update though.