So, I would expect that when you switch back to ISP from a vault running on VPN, that is quicker and the vault restarts promptly… so, that would suggest VPN delay is slower than the vault is willing to wait… seems not unreasonable - IP address should be available; if computer not connected then how long should it wait.
1 Like
Exactly,
This opens the question regarding unstable ISP connections, (dropping for a few seconds like a hiccup). If the vault doesn’t wait long enough and just closes, farmers wouldn’t know unless they are actively monitoring the vault. They have to manually restart it like I did.
I’m not a networking expert but online connections are never 100%. A vault should have some reasonable wait time (3-5 seconds) before restarting. But I’ll leave that up to MaidSafe and the devs to decide.
Again, I don’t think this is a problem, just an observation regarding Test 5.
1 Like
It’s a good thought… ISP connections can drop for a couple of hours… whenever an IP becomes available that should be used. Vault could warn perhaps that it can’t see the network.
2 Likes
Agreed,
That would be a useful feature in the vault manager.
1 Like
Pure speculation and just to note, I spotted something new that I wonder is the nature of the “destructive”.
In the vault there are obvious stats but those also appears in the launcher, which I wasn’t aware of.
One of the details stated is suggests as “Stats - User (Request/Success/Failure) - Get: 3/2/1” and then Put/Delete etc… that detail, as suggested, is number of “Requests/Successes/Failures”… and the fraction of Fail is from here ~32%… for both vault and launcher it seems.
That failure rate then I wonder is higher than would be wanted, so perhaps that is a reflection of some in built fail, tending to ‘destructive’… and the network ploughs on regardless. 
It would be interesting to see if that Get failure rate is reflected across the network, as I guess it’s a key indicator.
1 Like
interesting. I had no failures and all successes for my vault running on ARMv7 over WiFi and after a day was >600 GETs
4 Likes
And today I am running a vault on a “Pocket CHIP” which is a CHIP with LCD + K/B
Pic when it is starting, no chunks yet (when I took the pics)
and
17 Likes
I’ve only had it for a few days.
Wifi not bad but slightly less than other devices, but I still have it running on low power setting.
No browser pre-installed on the Pocket. Don’t expect desktop or new tablet speed, its not a quad-core 64bit thats in the new stuff. Just apt-get a browser and run it, but the screen is low resolution for useful browser work.
Not tried any of the other stuff i2p freenet etc. Only had it a couple of days 
and have other things to deal with. But saying that I am unlikely to try them anyhow.
You will find plenty of opinions to read in their forums https://bbs.nextthing.co/ or should I say old school name bbs rather than forum.
Be best to leave it there since this is test topic and not hardware 
1 Like
I see from http://vault-count.safenet/ we have 140 odd vaults online, there were only about 40 early on in the test, I wonder if the 100 is a real 100 or droplets from Maidsafe themselves?
I feel the participation rate is pretty low and maybe a little worrying if we need to get to around 3000 nodes to get to preferred number of nodes to humm along at for stable service.
anyone?
new site, bit pants http://www.weather.safenet/
tried to steal if from my favourite weather site: WunderMap® | Interactive Weather Map and Radar | Weather Underground
Al
2 Likes
@tfa how did you do the DNS entries estimate?
Forget counts given by vault-count.safenet given a few days ago. This program crawls the nodes by recursively getting the nodes in close groups. The problem is the initial set of nodes I start from:
-
a first version of the program was starting with nodes in my own routing table. The results were OK in a little local network of about 20 nodes. But this algorithm misses a lot of nodes in a bigger network.
-
a second version was starting with close groups of about 30 random addresses. The results were much better but were fluctuating
-
the latest version starts from 32 uniformly distributed addresses, which gives stable results and I am confident that there are really currently about 140 nodes in the network.
All this is about line labelled Current total nodes.
For the line labelled Locally estimated number of vaults I do an extrapolation from the distance of furthest node in my close group as explained here, except that I don’t subract a fraction of standard deviation. This value is highly variable and should be averaged on several vaults but I have only one.
1 Like
I count the number of SD in my vault having type tag 5, I divide it by GROUP_SIZE and I multiply it by the total number of nodes. This is an approximation that should also be averaged on several vaults.
2 Likes
Running a safe vault node on one of our servers for the hell of it. This machine is one that I’ve seen 50 Megabytes (not megabits) per second out of, and is equipped with an 8-core, 3.4ghz xeon CPU. Because of past use I can say that the routing/networks that it’s attached to are top class.
Does anyone want the logfile from it? I can make it available if it would prove useful to anyone.
Update: That machine has now moved 9GB today.
5 Likes
2 Likes
I’ve just been running my node for the past week or so. I haven’t been visiting Safenet sites or testing either the launcher or the demo app. This is not because I don’t want to, but because I don’t know if there’s any need to do so.
If there is I hope the Maidsafe team will tell us so. Until they do all I will do is run the vault and when new software versions are released I will play around with them for a bit.
Perhaps @frabrunelle can let us know in future dev updates / Test network releases what kind of activity is required to make this a successful test. I’m sure lots of us are willing to go the length if specific tests are needed at specific times, etc.
5 Likes
While testing is going on, wouuld it be possible to temporarily put something in the code to let us know some stats about the entire network. Stats like how many vaults are running?
Dido (20 characters)
The new client interface has an ETA of Tuesday and that’ll be enough for me to start a client up. I’ve seen the mock up but can’t wait to see it in action!
Maybe I get too excited over things like this but I can’t stop believing in what this project is attempting.
EDIT
As soon as messaging is being tested people will flock to the network. I know quite a few sticks in the mud who won’t test it yet til that’s available. Some peoples kids!
8 Likes
long time no see when maid launcher