what means : Routing Table size: 30 ?
I’ve now started 5 more instances:
-
Put the binaries in 5 different folders. along with a script to run each.
-
Edited their respective safe_vault.crust.config files to listen on different ports.
-
Added port redirections for the new ports.
Running these 6 instances, I saw a lot more “DUP ACK” and “Retransmission” packets but it has settled down somewhat to a much higher level of traffic than with one instance.
2 Likes
Very graphic! Thanks for that!
3 Likes
Can’t log in with safe launcher. Is it just too busy right now and I need to keep trying?
I’ve seen it mentioned the /tmp folder on linux hosted the data but where would I find it on a Windows 7 machine? I believe I am connected also!
1 Like
In Explorer do a search for a folder named something like “safe-vault.” followed by a 12 character identifier.
My setup came almost to a halt half an hour ago, with routing table sizes of only 1 or 2, so I rebooted and started just 4 instances. Now the routing tables are around 3 or 4 in size, the traffic is somewhat slower than before and there are almost no “error” packets.
1 Like
%AppData%\Local\Temp\safe-vault.randomcharacters
3 Likes
Can start a vault (many less errors in the first few minutes then before I went to dinner [not sure if thats good of bad]) however, launcher says “waiting for establishing connection to SAFE network” on every computer I try it on.
Side note, FFI had crashed while I was out. I thought that was a sleep error because I frequently get that on my laptop when it comes back, however, this computer never sleeps to protect my raid from spinning down.
I get the same response as @wes and others from the launcher, using either my existing credentials or if I click create a new account and then enter a new PIN.
From Wireshark I see that the launcher appears to be talking to IP address 64.71.168.202 which might be the proxy. I attach here a representative sample of the packets around the time I run the launcher. Sorry, long past time to hit the sack so that’s it for now.
1 Like
routing table down to zero
have two established with zen.co.uk node
last file chunks came in around 5.5 hours ago
should i quit and refire the vault or can we leave vaults running and they automatically rejoin your vaults when you and others restart them?
cheers.
Yes that vault can not retire nicely @abittner it has done it’s job now 
1 Like
while the vault runs fine on a x64 linux machine here, when I try the x86 version on a x86 linux machine, I get :
thread ’ < main>’ panicked at ‘Unable to start crust::Service IoError(Error { repr: Custom(Custom { kind: Other, error: StringError(“Failed to create TCP hole punch server”) }) })’, /home/maidsafe/.cargo/registry/src/github.com-48ad6e4054423464/routing-0.16.3/src/core.rs:285
stack backtrace:
1: 0xb768e69c -
Segmentation fault (core dumped)
Nice job. But it’s too technical for me. I am wondering if the general info of update could be translated into Chinese and posted on 8btc: ChainNode 链节点 - 元宇宙社区
BTW, we just hosted an AMA for Gavin last night, http://8btc.com/thread-31877-1-1.html and got over 10k views so far.
I am wondering if @dirvine would like to do an AMA on 8btc some time in the future.
10 Likes
Does not work in virtualbox or on metal. Host is connected to internet on direct connection. Here is the error on metal:
[reivanen@* safe_vault-v0.6.0-linux-x64]$ ./safe_vault
WARN 10:40:00.900703554 [w_result lib.rs:288] Failed to find an IGD gateway on network interface enp9s0 91.153.237.xxx. igd::search_gateway_from_timeout returned an error: IO error: Resource temporarily unavailable (os error 11)
thread ‘’ panicked at ’
===================================================
| Option evaluated to None: “This must be supplied” |
', /home/maidsafe/.cargo/registry/src/github.com-88ac128001ac3a9a/crust-0.10.1/src/service.rs:231
note: Run with RUST_BACKTRACE=1 for a backtrace.
You didnt set the port I guess. Read the beginning post of this thread. Set the value in your vault config file.
It was supposedly not necessary if you are directly connected. My computer is plugged in to the one port on the adsl box that offers direct connection.
You need to be defining the tcp port for your vault process.
Port forwarding is not needed but the config file does need to have the port the vault process should bind to (should listen to), It simply means that there is no software logic as of yet that would make the vault use a default or automatic or random tcp port. Therefore you must set it in the config file. Thats all.
Emphasis is on “and then”
1 Like
I’m getting the feeling SAFE will be VEEEEEERY popular in China.
On another note what do you mean by “direectly connected”? You mean just have a lan connection to the router? Because I have that and it still doesn’t work. The router isn’t mine so port forwarding isn’t an option. (Bloody borrowed internet.)
1 Like