I believe the testnet goals shift with what they gather from internal testnets.
5 Likes
Where are we at, then?
Will there be main3 tonight or can I relax and get on with other stuff?
8 Likes
OK, there is a new network up, but we’re going to leave it overnight, then do some testing on it tomorrow morning, and we will create a new thread.
Edit:
Just to say, the logs were pulled from main before it was taken down, and I think DavidR is going to dig in to them to analyze what went wrong.
10 Likes
Nice one, folks, we are iterating fast, and maybe this one was too big a step, but nevertheless, we have more to look at now, I am not convinced the failure was bad actually. So good all around.
One thing we have done recently is removing igd so nodes from home may mean nodes are directly connectable for a wee while (i.e. use a cloud hosting provider). That also removes complexity in port forwarding etc.
So let’s progress quickly in small bites and we will crack this. Thanks again for helping us build this beast. It really feels like we are building together when you test and fire in suggestions/comments and give us this amazing feedback we could never afford.
18 Likes
That never worked for my router anyway, and I suspect for many others. I get the ideal scenario of not having to fiddle with router configuration, but also think knowledge of how to open ports if you run a node is not such a big ask; surely it’s like that for most of the crypto projects (I know it was like that for storj for example).
3 Likes
Amazing how it can help to accelerate coding process! It’s a great co-pilot but still making mistakes.
4 Likes
Time to get the CLI2GUI back online 
3 Likes
I’m still on an island:
safe networks sections
Network sections information for default network:
Read from: /home/mav/.safe/network_contacts/default
Genesis Key: PublicKey(0d58..9171)
Sections:
Prefix ‘’
Section key: PublicKey(0d58..9171)
Section keys chain: [(PublicKey(0d58..9171), 18446744073709551615)]
Elders:
| XorName | Age | Address |
| 738018.. | 255 | 134.209.16.34:12000 |
3 Likes
I wonder if your node version is different from the ones @maidsafe launched in this last iteration? I’m not sure which version we should use now?
1 Like
totally agree with this. with complex problems it never hurts to reproduce any result several times. then take tiny steps from there. we havent had a public testnet in a while, whats the rush now?
if you had a delicious dessert in front of you, would you not try and savour every small little bite? 
4 Likes
I disagree. It hurts my enthusiasm and willingness to participate, at least if the frequency of testnets is once a week.
On the other hand, I would be willing to run the exact same testnet with same release a few times back to back, just to see if there is any difference. After the first run of each iteration, the next runs could also be a bit more “set and forget” for the team? It could be run with same or different parameters, to try to find out other bugs. I think this would make sense, because it seems to me that the big effort is to get the first iteration up in a coordinated way. Then, when that’s done, and everyone have the software installed etc. why not make another run, with similar or different parameters? Could be done even regularly so that if first iteration starts on Thursday or Friday, the next one could start on Monday.
1 Like
What stops any of us (if approved by Maidsafe), from running and participating in multiple parallel testnets (release version or bleeding edge etc), with different purposes simultaneously in terms of testing out a specific bottleneck (or confirming the issue doesn’t happen repeatedly) of the network between parallel testnets, whether its per day, week, or per month etc? I think this all depends on limitations of resources, not only to setup the environment, but the time to review/audit the logs repeatedly, as well as the current state of the code base.
At the end of the day, whatever Maidsafe decides in these early stages as the testnet, I am personally on board to help test, break, and verify their progress on the network. I feel they know the inner workings of this technology the best, and between their internal testnets and public testnets… this is whats guiding them on what criteria to test on each testnet. At some point, they may open up to parallel test networks to speed up testing and verification, and at same time, they may limit it to sequential test networks to gauge incremental progress as is. Lets wait and watch to see how it plays out 
.
At the same time, I love the fact that @anon26713768 and others are contributing their time to setup community testnets as and when they see a nice checkpoint in Maidsafe’s code base, worthy of validating and seeing if the network holds up.
8 Likes
But Safe is for everyone and unlike those projects the aim is for anyone to be able to run a node and earn SNT.
4 Likes
Wow, I have been offline offline for 24 hours an so much has happened. This main/main2 in one thread is very confusing. I suggest every testnet has its own thread and number, so let’s have thread (separated from the update thread) name for example Testnet #23 that will have corresponding network name test23 instead of main. This way we can run several testnets in parallel without confusion.
4 Likes
Sorry but that’s delusional; there will always be a barrier, whether that’s age, financial means or technical knowledge. In any case, we are still far away from the final product, and we all agree the barrier to entry should be as low as possible, but will never be non existent.
2 Likes
Morning! So (not enirely suprisngly), this one fell down.
I agree with folks on both counts: move fast and test things + wouldn’t it be nice to see stability.
I think as we go, we’ll likely try to have more experimental testnets run alongside some more stable code (on another testnet). The crunch there will be making sure we’ve tooling in place to make it clear what/where you are connecting, and if we can do that concurrently.
12 Likes
Is another one still expected today?
I’m poking at this just now. Waiting on some PRs to be review and merged. I think once we have this in, we can look at another limited size network (but one that should allow joins when appropriate, not in the free-for-all fashion).
10 Likes
Cool! Now I know to leave my seat for a while and return on the very edge of it later. 
7 Likes
Morning 
Single section, joins as needed, am I following?
5 Likes