Is there a plan for upgrading between incompatible versions yet or are we just trying to stabilise the network first?
I ask, as this will surely be a key feature (i.e. rolling out breaking changes).
2 Likes
If it’s incompatible (as in restricted), it’s intended to be separate. That’s the current situation.
So that network could upgrade to another compatible with itself.
Meanwhile we could have another network (stable) still ongoing.
(We just need some more checks on the release process flow to ensure we’re not building for one testnet from a branch restricting for another… if that makes sense).
Another thing will be working with breaking changes to a live network which will require some upgrade path handling (which is probably what you mean now i think of it).
We don’t have any of that yet, but we’ll be able to work it up when it’s worthwhile. Which will probably be soon…
8 Likes
My safe_cost_tracker script is still running. It’s nothing fancy. It just gets an estimate for storing a minimal sized record every 1 minute.
It just logs the output of the command in a file in this format straight from the output:-
Upload Run 6813
Sat May 25 08:14:32 UTC 2024
Logging to directory: "/home/ubuntu/.local/share/safe/client/logs/log_2024-05-25_08-14-32"
safe client built with git version: 16f3484 / stable / 16f3484 / 2024-05-09
Instantiating a SAFE client...
Connecting to the network with 49 peers
**************************************
Your current balance: 0.089998434
Transfer cost estimate: 0.000000080
Your balance estimate after transfer: 0.089998354
**************************************
I’d like to log the results in a better way and I’ve been Googling how to do so. I think I’ve got it now. Do we think XML or JSON would be better?
I think the only things I want to log are the time and the cost. No need for any of the other things captured from the command.
I’d probably log the time in the +%S seconds since the epoch format for future flexibility.
Are we XML or JSON fans for something like this?
Json is the lesser of two weevils IMO. 
If it will stay simple, CSV might be easier for other scripts to process.
2 Likes
I just shut down my ten cloud nodes and I’d like to know if anyone beat the storecost of Node 10, at 23 million SNT per chunk! 
6 Likes
Yeah, I can beat that by a bit:-
It’s a 9 with 16 numbers behind it.
It hadn’t earned since near the beginning and it’s one I shutdown a couple of days ago when the internet was a bit dicey here. Looks like 15 nodes is about all I can supply from home. 
3 Likes
I’m going to be bringing down the maidsafe nodes here today I think.
Thanks everyone who got involved here! All super helpful as ever!! 
11 Likes
This should be the case. The UPnP mapping is constantly renewed. This is logged as upnp::event::ExpiredExternalAddr.
3 Likes
Thanks for the advice. I’m now thinking it might be best as CSV though! Reason being that surely the best thing to do is load the output into a table in a simple database. Or even better - output straight into the table. Then it will be possible to publish a page showing the rolling average over the last day, week, month, etc. and other useful things.
And I’ve realised I definitely want to track the time taken to get the quote because it varies from a couple of seconds to tens of seconds and that could provide insights.
2 Likes