"community1" Test Network is Alive, Join Us!

bluebird · May 27, 2016, 10:39am

Just a reminder that this network is still operating and the plan is to upgrade the binary on the seed vault to the latest released version after the next official testnet (testnet4, early next week).

My testing on client_test_2 (the droplet network put up yesterday for app developers) shows that the newer binary probably does a better job of data retention (if we assume that the effect is not due to their having a slightly larger number of vaults.

The binary to be used in testnet4 will, hopefully, be just as good. That’s the one I’ll install on community1’s seed vault.

Please note that once that change is made, your existing vaults and launchers won’t be able to connect to it; you will need to upgrade to the same vault software and launchers (the ones to be released for testnet4).

rupert · May 27, 2016, 11:41am

hi all,

have added a second vault on: 185.16.37.156

if it would be handy to make it a seeder please let me know what else i need to do other than open 5438 tcp/udp

rup

bluebird · May 27, 2016, 12:26pm

Thank you! I’ll add it once we can confirm it is acting as a contact.

I ran a vault just now with only your IP in its config, and with no other vaults running on the LAN, and it just listened for a couple of minutes and then gave an error “address already in use” and then exited the program. I ran the test twice with the same result.

If your vault were accessible then I would expect this test vault to request a tunnel and it would join the network that way.

If it weren’t accessible then I would expect this test vault to just listen forever, with occasional restarts.

Could I ask you to confirm the config and the command you are using? And let me know your operating system.

Assuming Linux, the command should (at this point) be:

    $ ./safe_vault

(don’t type the dollar sign)

The config might be:

{
  "hard_coded_contacts": [
    {
      "tcp_acceptors": [
        "91.121.173.204:5483"
      ],
      "tcp_mapper_servers": []
    },
    {
      "tcp_acceptors": [
        "185.16.37.156:5483"
      ],
      "tcp_mapper_servers": []
    }
  ],
  "tcp_acceptor_port": 5483,
  "service_discovery_port": null,
  "bootstrap_cache_name": null,
  "tcp_mapper_servers": [],
  "network_name": "community1"
}

Let’s do that and then we can go to the next step.

rupert · May 27, 2016, 3:29pm

hi,

have tweaked the config as above so please have a sniff now…

rup

rupert · May 27, 2016, 3:30pm

ps. it looks pretty busy now so i think from what i can see it might be doing the biz…

r

bluebird · May 27, 2016, 3:48pm

I’m still getting “address already in use.”

Let’s check four more things:

Are you running the default binary for community1 (the link is in the OP). Not the one put up yesterday.
Do you see any entries in your routing table?
Did you start the vault without any flags (i.e., no --first)?

If the answer is yes to all three of those, then let’s try opening the firewall port (since my assumptions have proven innaccurate):

    $ sudo iptables -I INPUT -p tcp --dport 5483 -j ACCEPT

(on Linux)

Is the port directly connected to the Internet? I.e., if you are on NAT then you need to redirect the port to your computer. (I know Crust can do marvellous things, but let’s stick to what we know)

rupert · May 27, 2016, 4:12pm

lol…

running 0.8.1 as used in test3

routing table size is 16

INFO 15:59:15.124254498 [routing::core core.rs:1574] Node(54c7..) Added 296f.. to routing table.
INFO 15:59:15.124460981 [routing::core core.rs:414]  ------------------------------------------------------- 
INFO 15:59:15.124498712 [routing::core core.rs:416] | Node(54c7..) PeerId(042d..) - Routing Table size:  16 |
INFO 15:59:15.124533174 [routing::core core.rs:417]  ------------------------------------------------------- 
INFO 15:59:15.363691139 [routing::stats stats.rs:127] Stats - Sent 12500 messages in total, 34 uncategorised
INFO 15:59:15.364157643 [routing::stats stats.rs:130] Stats - Direct - NodeIdentify: 20, NewNode: 61, ConnectionUnneeded: 0
INFO 15:59:15.364377220 [routing::stats stats.rs:134] Stats - Requests - Get: 1790, Put: 1371, Post: 263, Delete: 0, GetNetworkName: 1, ExpectCloseNode: 7, GetCloseGroup: 324, Refresh: 5301, Connect: 4, ConnectionInfo: 694, GetPublicId: 0, GetPublicIdWithConnectionInfo: 0
INFO 15:59:15.364637388 [routing::stats stats.rs:149] Stats - Responses - GetSuccess: 1592, GetFailure: 4, PutSuccess: 480, PutFailure: 4, PostSuccess: 312, PostFailure: 0, DeleteSuccess: 0, DeleteFailure: 0, GetCloseGroup: 235, GetPublicId: 0, GetPublicIdWithConnectionInfo: 0, GetNetworkName: 3

used:

./safe_vault

to start server operation

routing to internet is pretty clean with no nat just a couple of routers between the server and freedom…

traceroute to bbc.co.uk (212.58.244.22), 30 hops max, 60 byte packets
 1  185.16.37.129 (185.16.37.129)  0.452 ms  0.337 ms  0.987 ms
 2  r1atm.skynode.pl (185.16.37.5)  0.943 ms  0.696 ms  0.650 ms
 3  te0-0-1-1.rcr11.b036814-0.waw01.atlas.cogentco.com (149.6.70.193)  0.979 ms  3.123 ms  3.110 ms
 4  be2407.ccr21.waw01.atlas.cogentco.com (154.54.37.57)  1.257 ms  1.567 ms  1.497 ms
 5  be2250.ccr21.bts01.atlas.cogentco.com (154.54.60.245)  14.225 ms  14.013 ms be2251.ccr22.bts01.atlas.cogentco.com (154.54.60.249)  14.230 ms
 .........

ufw (not using iptables) settings:


To                         Action      From
--                         ------      ----
22/tcp                     ALLOW       Anywhere
22/udp                     ALLOW       Anywhere
5438/tcp                   ALLOW       Anywhere
5438/udp                   ALLOW       Anywhere
9090/tcp                   ALLOW       Anywhere
5222                       ALLOW       Anywhere
5269                       ALLOW       Anywhere
22/tcp (v6)                ALLOW       Anywhere (v6)
22/udp (v6)                ALLOW       Anywhere (v6)
5438/tcp (v6)              ALLOW       Anywhere (v6)
5438/udp (v6)              ALLOW       Anywhere (v6)
9090/tcp (v6)              ALLOW       Anywhere (v6)
5222 (v6)                  ALLOW       Anywhere (v6)
5269 (v6)                  ALLOW       Anywhere (v6)

so as far as i can see i’m good to go…

rup

rupert · May 27, 2016, 4:18pm

have cleared everything down and restarted the vault… its picking up table entries and has reached 16…

now busy receiving data:

drwxr-xr-x  2 root     root     147456 May 27 16:17 safe-vault.bvFjwjnU1rTh

rup

rupert · May 27, 2016, 4:41pm

btw…

where are you connecting in from?

i’ve got:

.nl
.fi
.se
.de
.net
.ch
.com

nice to see such an international collection…

rup

bluebird · May 27, 2016, 5:00pm

Your port is not accessible.

You need to be listening on a port accessible from the Internet. As I mentioned, if you’re on NAT then you will need to redirect that port (with a rule in your router).

This is what I see when I telnet to the existing seed vault:

    $ telnet 91.121.173.204 5483
    Trying 91.121.173.204...
    Connected to 91.121.173.204.
    Escape character is '^]'.

This is what I see when I telnet to you:

    $ telnet 185.16.37.156 5483
    Trying 185.16.37.156...

rupert · May 27, 2016, 5:06pm

have telnetted it myself but was unsure if i should expect to see anything back…

will investigate further… and let you know when its sorted…

rup

bluebird · May 27, 2016, 5:13pm

I don’t know about UFW, but with IPTABLES in its default setup, if you reboot, you need to apply the port opening rule again (or set up saving of rules).

bluebird · May 27, 2016, 5:20pm

OK! I see you now!

I’ll add it to the config.

rupert · May 27, 2016, 5:24pm

hi,

yep… ufw was being naughty…

reinstalled and cleaned up rules… hurrah!

rup

bluebird · May 27, 2016, 9:43pm

Somebody’s doing a lot of restarts…

bluebird · May 28, 2016, 1:20am

Someone is running or trying to run the newer version of the vault released on Thursday, and IT IS CAUSING SERIOUS PROBLEMS WITH THE SEED VAULT AND THEREFORE WITH THE NETWORK.

Most of the log, SINCE THURSDAY, which has bloated to 15MB, consists of the line:

ERROR 23:59:34.966929639 [crust::connection connection.rs:416] Incompatible protocol version.

The log is being truncated, perhaps because of the flood of these error messages, and the vault has restarted at least once without any other apparent cause.

Please use ONLY the binary linked in the OP of this topic, until Maidsafe provides some way to simply ignore incompatible binaries rather than having an extended conversation with them.

EDIT: I’ve opened an issue on this.

bluebird · May 28, 2016, 2:17am

I have restarted the vault, just to confirm that my analysis of the situation is correct, and not a fault state in the vault.

Infining · May 28, 2016, 2:22am

That was me

bluebird · May 28, 2016, 2:25am

Are you 212...* ? (20 chars)

Infining · May 28, 2016, 2:27am

Don’t think that was my ip but I was doing some testing earlier today, including a ton of restarts.

Topic		Replies	Views
"community1" Test Network Has Restarted, Join Us! Development test_network	143	5870	June 29, 2016
Join the NEW Community Test Network (Running Now): "community1" Development	244	10619	May 21, 2016
Launch of a community safe network Community	426	14957	March 6, 2020
Community-run Testnet Info! Development	54	4804	May 13, 2016
Community1 test network 18-hour run Development test_network	27	3292	July 15, 2016

"community1" Test Network is Alive, Join Us!

Related Topics