MaidSafe Dev Update :safe: 21st April 2016

Development Updates
144

No the test is over and we have collected real valuable data. We will publish the results on Monday with a new test very soon thereafter. Basically we found improvements in

  1. Churn handling (we were way aggressive there) [almost done]
  2. Network segmentation prevention (this one was really important for us) [done]
  3. Huge improvements in mass network startup [done]
  4. Improve error/warning messages for early testers [partially done]

We will test again these changes on Monday so you will not have a wait to see the results.

We will explain these in more detail on Monday but this initial test was very very fruitful and you will see the results in the next few days. Very enlightening indeed and allowed the Engineers to focus on the important issues for stability and of course launching the network (our ultimate goal). So you will see a much more stable next test soon, but be warned it will be better by a long way but we will still kill it fast.

Thanks everyone for doing this, it really was extraordinarily helpful. More to come soon :wink: We will make a few small such tests, but now we can do it with a nice fast frequency. Very good indeed. I will sleep well this weekend. Beer tonight for sure :smiley:

34 Likes
145

Were you able to gather any stats on how many nodes joined the network in what timeframe? I’m really curious what type of adoption speed we have now, even before MVP. In short, how many techies jumped in how fast?

2 Likes
146

No we never got that info this time. We will run with a visualiser at times soon and that may show some more info like this. We will have heavy warnings when we do though and they will be short lived.

6 Likes
147

@dirvine will there still be a droplet testnet to build with? Or going forward strictly with the new?

148

These will be new for now. We should try and keep a test net and fixed net up, but hard as folks will join the fixed net with vaults I imagine. We hope this testing will not take long. For dev you can create your own mini network though. Start valts locally (wait till Monday) and they will find each other on your LAN.

It’s a slip really to kill the network like this too often, but the costs to keep several networks going is too much for us right now. We will solve that soon though.

6 Likes
149

Yepp, Thought to make own LAN network to build against, might be worth it.

Very exciting moment though, we connected on each others network yesterday.

8 Likes
150

Soon batteries recharged

1 Like
151

In a future release, will there be a linux binary compiled for ARM processors (raspberry pi)?

When I tried to run the vault on raspberry pi, it just returned an error. When I searched for the error, it seems this is because the binaries were not compiled for the ARM processor.

1 Like
152

ARM is not supported yet. Its in the pipeline, however, as of yet they’re focusing on win, x86*nix and macOS

1 Like
153

Looks like that will be addressed.

154

In my opinion, the release of the vaults doesn’t look good. How come? Is this not tested before it came live?

155

This was basically a “what happens when a mass of people all join ad the same time, and what does it break?” test. They’ll fix what they think needs to be fixed and we’ll try it again. Productive regardless initial result.

4 Likes
156

I couldn’t get it to work. But like they said, it was for the geeks out here to try it out first. Probably a better test as well instead of hundreds of Vault coming online at once. According to David they go some great insights. So that sounds great. I’ll wait for the MVP to join the network.

7 Likes
157

This was for the capable to help test vaults. Rather than for normal use. As @anon40790172 said [quote=“polpolrene, post:156, topic:8826”]
it was for the geeks out here to try it out first. Probably a better test as well instead of hundreds of Vault coming online at once.
[/quote]

4 Likes
159

Can’t get it to work at all.

TPGs-Mac-Pro:safe_vault-v0.6.0-osx-x64 tpg$ RUST_LOG=routing=info ./safe_vault
WARN 18:33:21.870952000 [w_result lib.rs:288] Error mapping external address and port through IGD gateway at address 192.168.192.1:80. igd::Gateway::get_any_address returned an error: Request error. HTTP error. connection reset
WARN 18:33:22.221132000 [w_result lib.rs:288] Error mapping external address and port through IGD gateway at address 192.168.192.1:80. igd::Gateway::get_any_address returned an error: Request error. HTTP error. connection reset
WARN 18:33:22.579385000 [w_result lib.rs:288] Error mapping external address and port through IGD gateway at address 192.168.192.1:80. igd::Gateway::get_any_address returned an error: Request error. HTTP error. connection reset
WARN 18:33:22.601859000 [crust::connection connection.rs:163] TCP direct connect failed: Connection refused (os error 61)
WARN 18:33:22.601893000 [crust::bootstrap bootstrap.rs:121] Error connecting to bootstrap peer: Connection refused (os error 61)
WARN 18:33:22.623253000 [crust::connection connection.rs:163] TCP direct connect failed: Connection refused (os error 61)
WARN 18:33:22.623272000 [crust::bootstrap bootstrap.rs:121] Error connecting to bootstrap peer: Connection refused (os error 61)
WARN 18:33:22.647629000 [crust::connection connection.rs:163] TCP direct connect failed: Connection refused (os error 61)
WARN 18:33:22.647646000 [crust::bootstrap bootstrap.rs:121] Error connecting to bootstrap peer: Connection refused (os error 61)
WARN 18:33:58.641725000 [crust::connection connection.rs:163] TCP direct connect failed: Connection refused (os error 61)
WARN 18:33:58.641739000 [crust::bootstrap bootstrap.rs:121] Error connecting to bootstrap peer: Connection refused (os error 61)
WARN 18:33:58.661668000 [crust::connection connection.rs:163] TCP direct connect failed: Connection refused (os error 61)
WARN 18:33:58.661688000 [crust::bootstrap bootstrap.rs:121] Error connecting to bootstrap peer: Connection refused (os error 61)
WARN 18:33:58.681862000 [crust::connection connection.rs:163] TCP direct connect failed: Connection refused (os error 61)
WARN 18:33:58.681894000 [crust::bootstrap bootstrap.rs:121] Error connecting to bootstrap peer: Connection refused (os error 61)
WARN 18:33:59.292067000 [w_result lib.rs:288] Error mapping external address and port through IGD gateway at address 192.168.192.1:80. igd::Gateway::get_any_address returned an error: Request error. HTTP error. connection reset
INFO 18:33:59.292212000 [routing::core core.rs:1231] Running listener.

Opened port in config and router. Even restarted router.

1 Like
160

The rest of the network is not up and running right now. The test only lasted a few hours. Bootstrap nodes have been taken down. Looks like some time next week (subject to change) we should have another test and it might last longer - depending on how the changes made from this experiment work out.

7 Likes
161

@Ross how about putting up a testnet status page people can monitor?

If so, I suggest that when it is live it includes a link to “how to join this testnet” as well as an expected duration / end time, the relevant vault & launcher versions, and a short description of the testnet purpose, plus summary of capabilities and limitations.

When it is not live, it can include an estimate of when the next testnet will begin ď if you’re really ambitious :slight_smile:, with suitable caveat!)

14 Likes
162

Yes, yes. Exactly.

So to put it in plain English for everyone to realize, right here, right now:

THE FIRST VAULT TEST HAS COMPLETED SUCCESSFULLY. DO NOT ATTEMPT TO FOLLOW THE ORIGINAL POST ANYMORE / I.E. CONNECT YOUR VAULT TO THE NETWORK.

Wait for the next update before continuing. (Unless you just want to mess around with this one 4 no reason.)

The next update, you will not have to jump through complicated hoops to connect your vaults to everyone’s! (I may be wrong on that. But everything so far has pointed to the fact that the next update will provide easy, instant interconnected Vaults for most people.) This thread’s update was a necessary albeit complicated step to test the waters for field bugs. Or water bugs.

5 Likes
163

@happybeing this makes a lot of sense - I will raise it at our morning catch-up tomorrow morning.

10 Likes
164

(safe_launcher-v0.4.2-win-x64 ) i downloaded it on my windows 10 from yesterday till now is still on ( Trying to connect with SAFE Network )

2 Likes