What did you change to join from your previous attempt?
I get the same error you were getting
1 Like
Haha, I donāt blame you or @Josh, you guys have been at it for a couple days now. Truly appreciated.
I am on a MacBook Air running BigSur 11.5.1
I open the terminal and run
curl -so- https://sn-api.s3.amazonaws.com/install.sh | bash
to download safe CLI and I see this in the terminal
curl -so- https://sn-api.s3.amazonaws.com/install.sh | bash => Downloading Safe CLI package from 'https://sn-api.s3.eu-west-2.amazonaws.com/sn_cli-latest-x86_64-apple-darwin.tar.gz'...
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 7486k 100 7486k 0 0 4302k 0 0:00:01 0:00:01 --:--:-- 4300k
=> Unpacking Safe CLI to '/Users/me/.safe/cli'...
=> Adding statement to '/Users/me/.bash_profile' profile to have Safe CLI binary path in the $PATH
=> Profile '/Users/me/.bash_profile' already contains a statement to set Safe CLI in the $PATH
me@Me-MacBook-Air ~ % safe node install --version 0.32.0
zsh: command not found: safe
me@Me-MacBook-Air ~ % node install
zsh: command not found: node
I tried node install without the safe command because that had worked for me in the past.
When I download sn_node 0.32.0 from the GitHub releases page and launch it I get
Starting logging to stdout
Error:
0: Cannot start node. If this is the first node on the network pass the local address to be used using --first
1: Routing error:: Configuration error: Network's genesis key was not provided.
2: Configuration error: Network's genesis key was not provided.
Location: src/bin/sn_node.rs:201
Backtrace omitted.
Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.
Saving session...
...copying shared history...
...saving history...truncating history files...
...completed.
[Process completed]
I donāt know how to add the genesis key/config file because as soon as I open sn_node it executes.
Iāve tried several different things that also didnāt work but this is what I was reading as the proper steps but being I canāt get er going that is likely not the case.
The other process I tried was downloading sn_node in the terminal and trying to run RUST_LOG=info ~/.safe/node/sn_node --hard-coded-contacts '["178.62.91.184:12000"]' --genesis-key
817193a6d3ad98e543cfcd0fbd5a5d9152d4f040205646509c3120f10fec268fa13ddb8199116ecf37fe8a12b3466342 --local-addr xx.xx.xx.xx:0 --skip-igd with my IP where the xās are like posted in the OP. See here.
curl -so- https://sn-api.s3.amazonaws.com/install.sh | bash safe node install --version 0.32.0
=> Downloading Safe CLI package from 'https://sn-api.s3.eu-west-2.amazonaws.com/sn_cli-latest-x86_64-apple-darwin.tar.gz'...
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 7486k 100 7486k 0 0 4097k 0 0:00:01 0:00:01 --:--:-- 4095k
=> Unpacking Safe CLI to '/Users/me/.safe/cli'...
=> Adding statement to '/Users/me/.bash_profile' profile to have Safe CLI binary path in the $PATH
=> Profile '/Users/me/.bash_profile' already contains a statement to set Safe CLI in the $PATH
zsh: command not found: safe
me@Me-MacBook-Air ~ % RUST_LOG=info ~/.safe/node/sn_node --hard-coded-contacts '["178.62.91.184:12000"]' --genesis-key
817193a6d3ad98e543cfcd0fbd5a5d9152d4f040205646509c3120f10fec268fa13ddb8199116ecf37fe8a12b3466342 --local-addr xx.xx.xx.xx:0 --skip-igd
error: Found argument '--genesis-key' which wasn't expected, or isn't valid in this context
USAGE: sn_node --hard-coded-contacts <hard-coded-contacts>
Edit: I donāt think Iām even getting sn_node 0.32.0 downloaded at all because safe, node, safe node install --version 0.32.0 any combination are not recognized commands.
1 Like
Sometimes you may need to restart your terminal after installing the CLI before installing sn_node.
How are you opening sn_node?
You only need enter the path to your node binary along with the genesis key and a ip from the config.
So try this in your terminal
~/.safe/node/sn_node --hard-coded-contacts '["178.62.91.184:12000"]' --genesis-key 817193a6d3ad98e543cfcd0fbd5a5d9152d4f040205646509c3120f10fec268fa13ddb8199116ecf37fe8a12b3466342 --local-addr xx.xxx.xx.xxx:0
Adding your ip at the end in place in place of xx.xxx.xx.xx keep the :0
If that does not work hopefully Chris can be of better assistance.
2 Likes
Havenāt tried this yet, Iāll give it a shot.
Part of my ignorance is manually setting Paths. Not sure how to do that yet. I see when I download things in the Terminal it says the Paths are set for command for safe CLI etc but commands never seem to work so far.
I amended my previous post to add my experience trying that but Iāll restart the terminal and give it all another shot.
Thanks Josh
2 Likes
similarly, what worked for me was using --local-addr 0.0.0.0:0
still waiting to get in, but at least connected to the bootstrap node:
ubuntu@instance-20210709-0856:~/sn_api-0.38.0$ RUST_LOG=info ~/.safe/node/sn_node --hard-coded-contacts '["178.62.91.184:12000"]' --genesis-key 817193a6d3ad98e543cfcd0fbd5a5d9152d4f040205646509c3120f10fec268fa13ddb8199116ecf37fe8a12b3466342 --local-addr 0.0.0.0:0 --skip-igd Starting logging to stdout
Oct 09 03:23:05.644 INFO sn_node:
Running safe_network v0.32.0
============================
Oct 09 03:23:05.648 INFO safe_network::routing::routing_api: 714e82.. Bootstrapping a new node.
Oct 09 03:23:06.804 INFO safe_network::routing::routing_api: 714e82.. Joining as a new node (PID: 637031) our socket: 140.238.207.60:38873, bootstrapper was: 178.62.91.184:12000, network's genesis key: PublicKey(0171..ec21)
Oct 09 03:23:06.806 INFO safe_network::routing::core::bootstrap::join: Sending JoinRequest { section_key: PublicKey(0171..ec21), resource_proof_response: None } to [(714e82(01110001).., 178.62.91.184:12000)]
Encountered a timeout while trying to join the network. Retrying after 3 minutes.
Oct 09 03:26:05.656 ERROR sn_node: Encountered a timeout while trying to join the network. Retrying after 3 minutes.
4 Likes
And if you run safe node bin-version in the terminal it says 0.32.0
I hope you get it sorted. Need to catch some zās. Good luck
3 Likes
This only works for me if I append --skip-igd
and then it times out but at least it connects.
2 Likes
I havenāt had any luck connecting in the past few hours since I successfully connected a while ago. It keeps timing out. This is with --skip-igd and --local-addr 0.0.0.0:12000.
On another note, I also tried without the argument --skip-igd (testing that feature out) with manual NAT rules disabled, and ran the following command line:
RUST_LOG=info $HOME/.safe/node/sn_node --hard-coded-contacts ā[ā178.62.91.184:12000ā]ā --genesis-key 817193a6d3ad98e543cfcd0fbd5a5d9152d4f040205646509c3120f10fec268fa13ddb8199116ecf37fe
8a12b3466342 --local-addr local_IP_of_VM:12000 --public-addr wan_ip:12000
It seemed to have configured the routerās UPnP properly seamlessly, where Int IP is the local IP of the VM running the safe node pid:
It then said bootstrapping a nodeā¦ Joining as a new nodeā¦ Sending JoinRequestsā¦ but eventually timed out:
I could see the state table on the router, and packets were sent on from the local VM IP to the bootstrap node 178.62.91.184, but stopped as soon as timeout was hit.
4 Likes
Yes, but no chunks.
(at least not in blobs directory. I suspect that small chunks may be stored inside db file)
Zero size of blobs directory is strange comparing with ~1.5GB from oct6 v1.
safe cat safe://hygoygyyd34jqo3phfarjez1kkph13zjnco7pocgttca77zuuzgddg3zieqso > 1.jpg fails with error
Error:
0: ā[91mNetDataError: Failed to GET Spot: NoResponseā[0m
Location:
ā[35m/rustc/c8dfcfe046a7680554bf4eb612bad840e7631c4b\library\core\src\result.
rsā[0m:ā[35m1897ā[0m
Backtrace omitted.
Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.
while safe dog works fine:
d:\SN>safe dog safe://hygoygyyd34jqo3phfarjez1kkph13zjnco7pocgttca77zuuzgddg3zie
qso
== URL resolution step 1 ==
Resolved from: safe://hygoygyyd34jqo3phfarjez1kkph13zjnco7pocgttca77zuuzgddg3zie
qso
= File =
XOR-URL: safe://hygoygyyd34jqo3phfarjez1kkph13zjnco7pocgttca77zuuzgddg3zieqso
XOR name: 0xce92e865bc2e08945e4a53792cdd22643ad830d18b31dede73b9863366f543ad
Native data type: PublicBlob
Media type: image/jpeg
Since it looks like there will be no useful data from this network anymore, I am uploading logs: Zippyshare.com - [now defunct] Free File Hosting
3 Likes
Just FYIs:
You shouldnāt need to have the genesis key or hard coded contacts in the node command if you already have the node_connection_info.config in the correct place (`~/.safe/node/node_connection_info.config).
No changes needed for CLI config once you have that too. (It should really be network.config or something to be fair).
4 Likes
First attempt was from my home PC, second was from a cloud VM
2 Likes
Thereās something strange going on with your shell setup here.
The CLI gets installed to ~/.safe/cli/safe and then the installer tries to add ~/.safe/cli to your Bash profile.
Based on other output though, it looks like youāre using zsh and not Bash, although I would suspect that zsh by default may read ~/.bash_profile, since zsh is a Bash compatible shell.
Regardless of what the reason is though, it looks like the location for safe isnāt on your path. Run env in your terminal and see what the value of PATH is. Post that, and also post the contents of your Bash profile please. You can use cat ~/.bash_profile to get that.
Assuming you have root access to your machine, you could also copy the safe binary from ~/.safe/cli/safe to /usr/local/bin/safe, since /usr/local/bin is almost always on PATH. Or of course you can run the binary directly by using ~/.safe/cli/safe.
As a sort of incidental point, Iām personally not so sure about the location weāve chosen as install locations for both safe and sn_node. Itās quite unusual in Linux to install binaries in hidden directories under the userās home directories. Those are normally for configuration files. I do understand though that thereās an advantage here in that you donāt need root access to put the binaries there. I personally like my binaries to go under /usr/local/bin when I have root access to my system. If I have to have multiple versions around Iād put them in /opt/safe-x.x.x and then symlink whatever particular version Iām interested in, into /usr/local/bin.
3 Likes
Yes I had wondered about that myself from time to time and agree that somewhere in /usr/local/ is def more āUNIXyā
2 Likes
Is anyone able to put/get I tried before leaving the house but could not. Heading back now and may take it down?
Iām guessing the reasoning was to avoid having to have root privileges for running the install and always having a location where you can write the binary to. It might be better though to offer both options.
2 Likes
Oh I understand why it was done but it was always, to my mind at least, one of these things that would get sorted before general release.
Other projects seem to survive and prosper while requiring root privs to install ā though of course in many cases that is āmaskedā by the use of apt or yum etc
2 Likes
Can you share the logs somewhere, please?
Iād like to grep them for the IPs I was using, see what if anything I can learn from that.
4 Likes
The network is down - logs here
5 Likes
env
TMPDIR=/var/folders/5w/5rzmzxrx0r15xw7rxns288cc0000gn/T/
__CFBundleIdentifier=com.apple.Terminal
XPC_FLAGS=0x0
TERM=xterm-256color
SSH_AUTH_SOCK=/private/tmp/com.apple.launchd.SuoxikV1ke/Listeners
XPC_SERVICE_NAME=0
TERM_PROGRAM=Apple_Terminal
TERM_PROGRAM_VERSION=440
TERM_SESSION_ID=F2CF8EAB-AE90-429F-A743-F217FE04AD9F
SHELL=/bin/zsh
HOME=/Users/me
LOGNAME=me
USER=me
PATH=/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin
SHLVL=1
PWD=/Users/me
OLDPWD=/Users/me
LANG=en_US.UTF-8
_=/usr/bin/env
Your sorcery has worked @chriso!
Welcome to Safe CLI interactive shell!
Type 'help' for a list of supported commands
Pass '--help' flag to any top level command for a complete list of supported subcommands and arguments
Type 'quit' to exit this shell. Enjoy it!
The ~/.safe/cli/safe command got the interactive shell running which has always been the only way I can get things to work. Thank you! I am keeping that command written down for later use. Looks like the test is over again but that is a huge help to me personally.
Edit: safe node install --version 0.32.0 still not working for me though. Hmm.
I just get
Format of the config file at '/Users/me/.safe/cli/config.json' is not valid and couldn't be parsed
Up to you whether there is any further digging required. Anything I can do to help get over this hump for others and round things out Iām happy to help but if itās not a good use of time, no worries either. Iāll prolly start learning a bit more about Terminal so I have a better grasp. Thanks for the help so far. 
1 Like
Yeah, so you can see PATH is not being modified by the installation process. Iām guessing now youāve copied safe to /usr/local/bin? The shell can then find safe because that location is on PATH.
Based on what I can see on Google, zsh doesnāt read ~/.bash_profile by default, and thatās why PATH isnāt being changed. Iām quite surprised we never ran into this before though. Iām pretty sure @StephenC has tested the install script on macOS fairly recently and it worked ok.
If I was you, I would just delete that ~/.safe/cli/config.json file. Itās quite possibly incompatible with what we have now and Iām not sure thereāll be much in there youāll be looking to keep.
Edit: @Nigel
3 Likes