All is in the title
Sadly still more information needed as there is literally an unlimited amount o ways how this 3 hostnames are “given”. How does the output-data of the node look like?
You are right jan
end of Function node :
items[0].json.myHosts = myHosts; return items;
myHosts is type [:string]
Then it could look like this:
Yes, that works jan, so a single exec can launch several command sequentially or in parallel
But here I need to analyse the command result (stdout) on each host or partially, depending of previous host’s result, am I clear ?
In fact, a function where I could fork (promise for ex) is fine, but the Function node is not made for that I think … Or, give to Exec command a very long command with if, else etc… and get the final output for the nex node
Am I right ? An other way ?
Yes, that works jan, so a single exec can launch several command sequentially or in parallel
Yes
But here I need to analyse the command result (stdout) on each host or partially, depending of previous host’s result, am I clear ?
Sure, if you have to do that you can do that.
In fact, a function where I could fork (promise for ex) is fine, but the Function node is not made for that I think … Or, give to Exec command a very long command with if, else etc… and get the final output for the nex node
Not sure I understand. But it is possible to return a promise in the Function-Node. Apart from that, can you do literally anything in the Function-Node. You can even use any module that is installed as long as you configured n8n accordingly:
If the code exists … jan do you have an example of a Function node usind the Exec-command node ?
That is not possible to use one node inside of another. What is possible is to use “child_process” in the Function-Node. Here the code of the Execute Command-Node:
https://github.com/n8n-io/n8n/blob/master/packages/nodes-base/nodes/ExecuteCommand.node.ts
or even simpler here some examples on stackoverflow:
Ok thanks jan!