SAFE Network - Test 16


Seems to me that MaidSafe screwed up big time by taking Alpha 1 down at exactly the same time as its apostles were spreading the gospel around the world. Would-be converts will have arrived at the site, only to find that there was nothing to see in terms of a live working network. Great way to kill enthusiasm. I know there were some problems with Alpha 1, and it was a bit out of date, but surely something is better than nothing?

So yes, as others are suggesting - have two networks. One based solely on MaidSafe droplets for external developers and one for internal and community testing that is not expected to be stable and can be refreshed every week, or whatever.


Yes that is true, however Alpha 1 was a droplet based network and digital ocean had caused many nodes to stop for some internal maintenance. So there was data loss on Alpha 1 and it would need refreshed. It may have been an idea in retrospect to ask folk to start Alpha 1 again. Alpha 2 nodes would have probably recovered that data. Anyway there is a lot of good info in this thread for us to consider. ofc we all know the ultimate answer really, turn this apparent threat into a good thing, so defeat these kind of attacks with the real system with the likes of node age etc. possibly remove tunnels at all (radical) and then the attacker need to really work hard and spend cash to help testing. That would be the best.


One good thing about the attacks is it shows people want to, and will try, to damage the network and as fixes come into place these open tests will be properly tested, against real aggressors. :slight_smile:


@frabrunelle :kissing_heart:


We have to remember there are economic actors at play. Sell high, then attack network, then post negative stuff on forum, then buy back. Rinse and repeat over the hype cycle.


Yes, fair enough. I understand the circumstances were out of your control and that the timing was unfortunate.Would it still make sense to restart Alpha 1 again do you think, just to have a live network for people to try?


Alpha 1 would likely be attacked too.


I don’t have all the insights @maidsafe has, but my 2 cent below:

attackers will attack, no matter if it’s a new coin (51% attack), new idea (DAO) or Bitcoin/ethereum (spam attacks). There will be blood :smiley:. MaidSafe is publishing stuff out in the open where you can compile your own Vault/Client whenever an update comes out. Quite cool, but this leaves room for attacks:

  • The attacker could remove the 1-vault-per-LAN restrictions and start a number of them in the cloud somewhere. He/she could easily run 20 to even more Vaults on 1 virtual machine I guess. So having 3 cloud instances could provide 60 nodes, which could kick ass when causing “evil churn” as the network is only 156 nodes.

To prevent this, release all your code for Vaults in the open except for a few small lines of code that prevent external compiled Vaults to join TEST xx. That way people have all the features, can build local networks, test stuff out but aren’t able to join a running TEST as the binaries provided by Maidsafe have 1 little trick which isn’t shared with the community until the test is over. Called it a second “hidden” networkname. Could be 1 key in the binary which is shared with the community after test is over.

And now about the Alpha’s…

  • The Browser with Authenticator are close.
  • Current Vaults are way better than the ones used in Alpha 1.
  • There’s an email App on the way
  • Mutable Data is on it’s way
  • Node.JS is on it’s way

So Alpha 2 could become:

  • A MaidSafe controlled network (just like alpha 1) spread over 3 or 4 cloud providers for more security.
  • SAFE Browser with Authenticator.
  • Email App that uses your public ID for email exchange.
  • Webhosing App for uploading Safesites.
  • Mutable Data.

This would open up a “public network” again where App_devs can build great Apps. It would be a very user-friendly experience with the Browser and Auhtenticator. The email_app would bring something new.

I wouldn’t mind waiting a few more weeks for this one. And when it’s live we could have some more TESTs. Is stuff working out fine? Allow some invited users to join Alpha 2 with some Vaults from home etc.


except for a few small lines of code that prevent external compiled Vaults to join TEST xx …

But what happened to ‘reverse engineering’ practices? If we’re to take future attackers seriously they would be able to circumvent this too.


Yes, probably, so should be 3 little tricks in 3 different places in the binary. I’m not een expert on this, but maybe a second hidden network name like I proposed, in a complete different module a function that asks for a hash of this value. And on another place a function that uses that value as a private key whenever another Vaults asks to sign something. so you would join the network as a Vault, other Vaults ask for the hash of that key, a few minutes later several nodes provide you with a random message they want you to sign. Like I said, I’m not an expert on this, but something like this could be helpful for some testnetst maybe?? Until we see the datachains and node_ageing in place???


Maybe if these are short tests so one is hindered for a while   ;o)

But seems a gliding scale to all at once just giving out binaries, no honor in that.

Better go for the real thing where the design stays intrinsically sound.
If theoretically and practically possible   o/o


Who cares about “honor” (sic) and lofty OSS ideals, purity blah blah blah at this time?.
There is a job to do, an engineering solution is required ASAP. Your “intrinsically sound” solution will be applied in due course.
Right now we need a workable - and nothing more than “workable” kludge to allow the MaidSafe and community devs to get on with developing apps.
Stallman-esque purity (which I fully subscribe to) can wait for later.


Oh boy; no not my solution, me, I was referring to the course MaidSafe took all along, main characteristic being, open system. Could be honor does not get equated to engineering merit much but it should, as in: do not come up with a wanky network which may only play into the cards of governmental and/or economic powers?


Would it be reasonable to assume that Alpha 1 was at all the time susceptible to the same attacks that we see Testnets 15 and 16 subjected to?
I can only speculate what would have the consequences been had it been attacked.


I like this idea. Personally, running my own vault once in a while, to prove to myself that we have something that will someday work in the real world, is valuable and important. However, I don’t need that confirmation on an ongoing basis.

It is important that we have a stable network for developers and others to play with, that is not vulnerable to sabotage by short-sellers, governments, and wanna-be hackers who mistake their ability to sabotage an incomplete network with evidence of their hacking genius.

I would trade community run vaults for a maidsafe run, invitation accessible, alpha network, since that arrangement allows maidsafe to harden the network on their schedule, rather than the schedule of some random script-kiddie.

If we want to do some community run test networks occasionally to test any hardening or NAT traversal that maidsafe has chosen to do on their own schedule and not an attackers schedule, I’ll be there with bells on to run it and help test. Until then, an invitation token style alpha network with all vaults run by maidsafe would be my preference, so we don’t slow down development.


I have modified my constellations site with the possibility to display the sections prefixes in the galaxy itself.

I also wanted that people could locate their own vaults in the network by entering their ids but that doesn’t work yet. I will be able to work on it next weekend.

Here is a screen shot with the new feature activated:

(there are currently 151 nodes and a section split has just occurred 001 => 0010 + 0011)

I was unable to reconnect to the same account (because of CoreError::GetAccountInfoFailure error). So I had to recreate an account with a new site name (safe://constellations2). safe://constellations is still available but is frozen at 2017-04-09 19-34-48 UTC.


No idea how you’re doing that but it’s so cool. :smiley:

I wonder that detail might be interesting to see evolve - I don’t know if it’s practical to retain the detail and run into a .gif motion.


I have received a PM asking what package I was using to build the graphical representation. I reply here because the answer may be of interest to other people.

The site is coded in raw JavaScript and I don’t use any package. You can read the source code by activating Beaker’s developer tools. There are few comments and code has been split in many js files to remain under 3KB limit. The dynamic part is defined in json files.

For graphics, I use canvas element. but last year I was using SVG (in test networks 5, 6 and 8). I did the migration because I intended to use utilities based on canvas to convert the page to mpeg or gif files. But finally, I opted for Screencastify extension when I generated videos for YouTube.


Guys how is #16 running? Is the problem of testnet 15 solved?



Hi Everyone,

I’ve been interested in maidsafe for more than a year but only realized recently that that active public testing was happening. Two days ago I requested an invitation and yesterday installed all the apps as described here, it all went smooth as butter. I published a small web site with one picture safe://tpirate/4tee2/ I also started a Vault and it has been running for about 24 hours.

If this is not the right forum topic let me know but…I’ve looked all over the forum site and can’t find the information I’m interested in.

  1. What are we (testers) supposed to do now? Is there an organized list of tasks we should be doing? Or is the test network just supposed to be up for the devs to monitor?

  2. Up to now I’ve left everything running. But say I have a power outage? Anything special I/we need to know to reconnect? Related, say I have to update my OS (I’m on debian 8 Jessie) what are the protocols to stop the Vault, do my thing, and then re-up the Vault? I’d like to test unloading and reloading the Vault. What are the recommended procedures and/or is there documentation for the ins and outs of running a Vault?

  3. What needs to be backed up? For instance, say I need/want to install all the apps on another computer not running a vault, do I need any config files/folders that will perform the magic? Or all I need is my secret and password?

  4. Any other relevant information for us testers?