It seems sudo strikes again. When starting normal (without sudo) these dirs are created in the user space (in /Users/yourname/) that’s where data will be etc. So everything in the dir is available to you.
Now if folk do sudo, it means run everything as root (the devil user form hell who can do anything, and let you get hacked etc. etc.). When running as root it’s not always sure the users dirs will be available, they usually are and then the sudo user (root, delivery …) can put stuff in there and make it not available to the user.
I would advise as many folks as possible not to use sudo unless they are trying to run system services. Those should not run in user dirs but in /etc /sbin/ /bin etc.
I think sudo is causing hell in this thread to be honest. It maybe best to remove all the safe dirs (which you may need to run sudo to do, just cause root may have created them) and reinstall the safe network stuff again from scratch.
I think the difference between root user and normal user instructions are causing issues fro sure and it’s totally understandable why (cc @chriso )
Most services need root to install, then run as their own user. This seems to be the normal way systemd is used.
We currently seem to have a hybrid of this, with systemd being used to run a node as a local user. Maybe we should either just run an executable as a user or install it as a systemd service using root, then run it via a separate user?
Edit: docker is also an alternative if it should be launched by a user, but in isolation. Either way, I don’t want nodes running under my regular user account (in case an expoit leaves my files vulnerable).
Buttt, if we’re talking about service definition… It seems like one potential bug in our approach (that there’s a fix for into main already) was that we stuck the WHOLE network-contacts file in there. Which is too much for some service managment apps… So that might be what’s going on there?
So I am back at giving Launchpad a shot after doing sudo and non sudo safenode-manager reset. You were right! there was existing nodes under sudo from previous attempts.
So far still no dice but if I scroll up in the Launchpad I see this
/Users/me/Downloads/node-launchpad ; exit;
/Users/me/.zshrc:source:1: no such file or directory: /Users/me/Library/Application
On 2013 iMac running Catalina but you may notice it cuts off at /Application Support for some reason, not showing the entire path. I’ve noticed it do this before but I don’t know why or how I got around it in the past tbh.
As far as the directories go just recently I got slightly further by ensuring I had necessary permissions with chmod +rw ~/.zshrc
Whether that makes sense or not, I’m not really sure as I’m still noobing with CLI over here.
As an aside, I’ve had zero issue with Launchpad on newer hardware and software across a couple laptops and has been super easy! Just this old dust collector giving me trouble.
At the moment I have 22 nodes running, and it seems to be going smoothly.
I think the limiting factor will be my routers NAT table - or the amount of connections in general. I wonder how this relatively idle state of the network compares to more busy network regarding the connections? Do I dare to try to increase the amount of nodes, or should I leave some room for more busy times?
Cool. I have FRITZ!Box 6660. I bought it just a couple of weeks ago, and in one testnet is seemed to tolerate somehting like 70 nodes, but later not that much. And even in the beginning that may have been an illusion.
I bought the box because it got good reviews in general, and especially as an replacement for the one my ISP provided. But it seems that consumer opinion of “general goodness” is not very good indicator of the goodness regarding our use case.
It would be interesting to gather stats on nodes per router and what the spread of said routers are. At 30 nodes, that’s only 60gb max. I wonder where that is for the total storage vs decentralizion balance?
The common complaint about my previous one, a Sagemcom, was that it’s connection was unreliable. And that’s for your typical user, let alone me. The speed and wifi range was OK though.
That’s with you anyway, it’s pretty good, I finally started 5 nodes yesterday but still safenode-manager as well as Launchpad show: Storage Allocated 0 GB, Memomry Use 0 MB and TNE 0 nanos, and the computer is terribly overloaded and keeps breaking my Internet connection, after stopping the nodes or removing them, the computer returns to normal. The nodes even though they only started up yesterday have been consuming resources and energy since they were added which is almost since the beginning of the Beta, unfortunately I can’t get any answers to my questions.
Sorry to hear that. Yeah, it’s tough since we all have our own hurdles and the team have so many people to answer to and so much more work and on a timescale no less.
I’m sure there will be a fix that will come down the line making it more easily accessible for our edge cases. My working nodes still haven’t earned, it’s also just the very beginning of wave 1 so don’t feel like you’ve missed out of much!