all these people with secret providers they are not sharing info about its not fair on the community keeping secrets what if we all want to run 25k nodes on a dedicated
People will be able to add nodes if they want to, but if so, it would be for altruistic reasons to help the project; there wonāt be payments. At least as far as Iām aware.
It would just be when the network becomes available. Iām not sure off the top of my head, but, it will be before the end of April.
I am hoping to be able to look at making it easier to distinguish between this network and the main production network. Might involve creating āchannelsā with some of the tooling, or something like that. Not completely sure.
Will be amazing to see it working after all these years. Hopefully there can be some testing on request for us to overload machines on purpose so the shunning can be dialed in properly.
I think there is phantom nodes in the quoting. That is the routing tables are not updating properly and nodes that are restarted when no port forwarding & not setting the port numbers is being used will result in a singular node appearing as 2 or more nodes in the network. Thus if a 10K node machine is restarted 2 times the 10K nodes could have 15K to 30K appearances in routing tables of nodes connecting to those nodes.
One potential reason why not enough quotes error occurs. Trying to get a quote from a phantom node doesnāt get a response.
And definitely they would be using swap. I have swap on my SBCs and they work fine. Swap is on nice fast SSDs and like @aatonnomicc I have 400 nodes running fine and room for more and the SBC is still very responsive and they get emissions
I am tending to what @Southside is saying, that until we have uploads working satisfactory that emissions should be drastically reduced. Lets make up the loss in emissions when uploads are working great and spread the make up in emissions across the remaining time left (approx 12 years)
Becoming silly to have such high emissions while we cannot upload anything using versions later than ant 0.3.8
Itās the hard drives that are the issue I am trying to point out the hetzners I put in to melt down was nearly idle on cpu and only half ram used. It hard drive io times that are the killer once there are to many nodes.