I had to restart my server a few minutes ago. The community1 network was down and the vault refused to run after stopping and starting it (no log activity at all). Strange! Rebooting the server has it up and listening and one other vault (mine) has connected to it.
I see only one entry in my router table. Could someone else with a vault please restart their vault?
Looking at the plot, I see that it was exactly two hours since I started the network and it “flatlined.” Veeeery interesting. I wonder if the devs inserted a timer, a poison pill as it were. My curiosity (apart from killing the cat) drives me on.
Then again, these vaults don’t do much unless there are seven of them. With one entry in its table, it is just listening and doing nothing else.
I’ll put in a cron job to restart the server every 90 minutes.
Not sure what the changes on the backend were, but I was running 7-10 vaults last test and for the community network no problem. This time around, while I’m searching for hosts with a single vault, my computer becomes unbearably slow. I guess its the blocking mechanic behind their connections (I know they said 2 threads per connection, and I have 2 real and 2 virtual adapters its trying to connect through). I hope that when MIO gets in there that searching for new peers will be streamlined.
I tried to connect but no success, is the IP 91.121.173.204:5483 still valid?
Edit: it only says:
INFO 10:58:48.894992946 [routing::core core.rs:1211] Running listener.
EDIT2: yay it’s working, I just made a mistake naming the config file
This is what it look like now, after renaming the log and restarting the vault:
Running safe_vault v0.8.1
=========================
DEBUG 08:58:28.262301733 [crust::service service.rs:107] Network name: Some("community1")
INFO 08:58:28.264125699 [routing::core core.rs:1211] Running listener.
INFO 08:58:28.264350954 [routing::core core.rs:564] Node(593c..) - Started a new network as a seed node.
DEBUG 08:58:28.265894520 [routing::core core.rs:1184] Node(593c..) Finished bootstrapping.
Notice the times. Nothing was being added before I removed the log, despite restarts. What’s with that? Shouldn’t it at least have a line about restarting?
EDIT: Ten minutes later, there’s one, new entry:
Running safe_vault v0.8.1
=========================
DEBUG 08:58:28.262301733 [crust::service service.rs:107] Network name: Some("community1")
INFO 08:58:28.264125699 [routing::core core.rs:1211] Running listener.
INFO 08:58:28.264350954 [routing::core core.rs:564] Node(593c..) - Started a new network as a seed node.
DEBUG 08:58:28.265894520 [routing::core core.rs:1184] Node(593c..) Finished bootstrapping.
DEBUG 09:11:38.190133829 [crust::connection connection.rs:610] Stats - 0 connections to 0 peers - direct: 0, punched: 0
DEBUG 09:11:38.261190820 [routing::core core.rs:1490] Node(593c..) Accepted client feb3...
DEBUG 09:11:38.311322455 [routing::core core.rs:1875] Responding to client Client { client_name: feb3.., proxy_node_name: 593c.., peer_id: PeerId(5ea2..) }: GetNetworkName { [PublicId(name: 593c..)], PublicId(name: ae63..), MessageId(ea51..) }.
DEBUG 09:11:38.370139649 [routing::core core.rs:1535] Node(593c..) Handling NodeIdentify from ae63...
INFO 09:11:38.370444131 [routing::core core.rs:1574] Node(593c..) Added ae63.. to routing table.
INFO 09:11:38.374619902 [routing::core core.rs:414] -------------------------------------------------------
INFO 09:11:38.374828282 [routing::core core.rs:416] | Node(593c..) PeerId(fecb..) - Routing Table size: 1 |
INFO 09:11:38.374983287 [routing::core core.rs:417] -------------------------------------------------------
DEBUG 09:11:38.375104445 [safe_vault::vault vault.rs:262] Vault connected
Something very weird is going on: my account, private and public data that I uploaded to the “community1” test net (2-3 nodes in routing table) appear now to be online on the official testnet (~50 nodes). I even thought the official testnet was already down… any explanation for this? Did I miss something_ Do I have to delete any cache?
EDIT: or is is possible that the two networks somehow merged? (sounds implausible…)
EDIT2: SOLVED: my launcher was connected to the official testnet…
Possibly you have a vault config that points to the community testnet and perhaps a launcher config that points to the main testnet. So your vault is on one and the launcher is on the other.
I see, I was not aware that the launcher has a separate config (but it makes sense when i recall that I don’t need a vault to use the launcher…) Thanks for the quick reply!
I am running everything on ubuntu. Some hours ago my vault was not able to connect to community1, so I switched back to the official testnet. It turned out that my launcher was on the official testnet all the time (see the posts above). It seems that the testnet is much more stable now than it was last evening/night (European time). I had the impression that everybody thought it was down, which seems not to be the case. When it is down I will switch back to the community net, of course (btw, what to write in the launcher config file for the community net? the same as in the vault config file?)
I’ve been out all day and just got back. Restarted the vault, since it was only showing one entry in the table.
@dirvine Is there anything in the current version, by design or accident, that would stop it from functioning indefinitely? It fails quicker than the previous community testnet on version 0,8,0 (connects faster though)? I have found twice now that deleting the log and restarting gets it going. Why would an existing log file make any difference?
In case it was failing because of the way I was invoking it in the start script, sending STDOUT and STDERR to /dev/null (a bit of a reach, since who ever heard of /dev/null getting full), I have redirected them to another log file:
$ cd /opt/safe_vault2/
$ ./safe_vault --first >> big.log 2>&1 &
