Hello, @HeptaSean
I only downloaded cardano node-1.35.4, and db as @lauris guided.
After that, I did as I showed via above image.
So I can’t understand what you are saying now.
If you do not start the node, there can be no socket the cardano-cli can connect to.
You must have chosen a way to start the node as you already showed us logs in your second post above.
We can’t know which way you chose there. You could start it directly from the command line or with a systemd unit or from a script (that is then in turn often called from a systemd unit). Only the admin – you in this case – can know what was decided after considering all alternatives for that specific machine.
Anyway: Where the socket path is created is decided by a parameter you give to cardano-node when you start it. Since we can’t know which of the ways you chose to start the node, we cannot know where the socket will be created and if cardano-cli does not find it, because it was not created by cardano-node up to that point or because you gave a wrong path.
If you just downloaded the database, it is totally plausible that the node takes a while to start, before it creates the socket, since it validates the database first. You should be able to see that in the logs.
I have downloaded a node via cardano-node run ./start.sh command.
But it has token over 15 days, so I deleted all cardano folder.
And downloaded new cardano-node version and db.
So I didn’t start any cardano-cli , node command yet.
Where did that start.sh come from? It’s not part of the standard distribution as far as I can see.
Downloading the DB can save you a large part of the sync time, but you still have to start the node, so that cardano-cli has something to connect to.
I made start.sh that is including this kind of content based on my folder structure.
So do I have start “cardano-node run” command now even if I have a downloaded db ?
Yes, of course, the download just saves the sync, not the running of the node.
How can I start a node ?
If I run above node-install command, the db will be overwritten.
I can’t know how I can do.
For example with the start.sh script that you just showed?
If you unpacked the downloaded database correctly, it uses it and does not overwrite it.
I can’t reinstall node db again, because it will take over 15~20 days. or forever.
This is why I downloaded a database as @lauris 's guide.
Now I am finding how to run this node and db.
First of all, you should know that the socket file will only be created once the node is running (not during startup). Second, during initial startup (the first time), it will sync the entire blockchain. On a typical Internet connection this took 24-48 hours last time I did it, which was in February or March. Longer now.
Third, I recommend that you turn on the logging in JSON format, and we have a tool that scans the logs for error codes and notifies you - called CodaEA. AlexD and a few others are using it - it scans the logs, reports the error to our cloud service, and looks up any community analysis of the meaning and solutions. We are in pre-production testing with CodaEA, but if you are having errors, certainly we can look them up and post the solution.
The log screen shots you are showing aren’t showing any error codes.
You can access the software on our Github - the docs are here CodaEA/CodaClient_CLI.md at main · info-tpr/CodaEA · GitHub. Scroll down to Text/cardano Specs, there are instructions for configuring CNODE to output JSON logs. Definitely message us on Discord if you have questions!
hmm…
It seems there is no way for me to implement this downloaded node.
@lauris , sorry, but could you help me based this snapshot db?