How to understand why the GET request is not processed?

Question

Computer networks

The task is as old as the world: there is a device with a web interface from which you can re-initialize ports that have not started on it. This is done by clicking on the button that sends the form, a GET request of the form: 10.0.10.26/DeviceCommand?param=0¶m2=reinit&pwd...
You need to do this on a schedule until I understand why the ports fall off. To do this, I wanted to create a task in cron like: curl -I 10.0.10.26/DeviceCommand?param=0¶m2=reinit&pwd... However, the request is not processed.
That is: from Firefox'a everything is all right. From the private session, too, everything is in order (I see in the logs). But with curl, lynx and phantomjs, the request does not work, although I get the html of the page and http 200 ok.
The question is, what is the difference between these tools for an embedded industrial hardware web server?
PS Authorization is disabled.

Answer 1

the button sending the request - part of the web interface?
maybe add some user-agent for curl..

Answer 2

Yes, part of the web interface. That's why I put phantomjs in order to press the button from it.
I also thought that this is the only obvious conclusion. But what difference does it make to a simple web server? In any case, I will try to eliminate this point. Thanks for the advice!
The idea just came to mind: to go firefox'om from the script. Tried: "c:\Program Files (x86)\Mozilla Firefox Private\firefox.exe" http^://10.0.10.26/DeviceCommand^?param=0^¶m2=reinit^&pwd=^&authorized=1^&config_path=^ &createName=^&createParams=
Reinitialization doesn't happen either. Obviously, it's not the client. Looks like wireshark is needed.