Has anyone tried running a script against the agent via the API?

[AshleyJackson]

The swagger docs are limited in explaining if any parameters are required for the below api call

In Postman it looks like that specific call (with and without any other paramters) just spits back a 500 Server Error

And on top of this, I can't find any live logs of this error happening
(NOTE: I am running in docker, but I don't think that should effect the api calls as other api calls work fine.)

[wh1te909]

Please read https://docs.tacticalrmm.com/functions/api/ you need to open your brower's dev tools, perform the action from the web UI and you will see exactly the payload you need

[saulens22]

The API documentation currently is very crude... Would it be possible to list all API endpoints in documentation? It should be possible using django-extensions command show_urls.

https://django-extensions.readthedocs.io/en/latest/command_extensions.html?highlight=show_urls#command-extensions

[wh1te909]

enable swagger and you will see all the endpoints

[AshleyJackson]

Hi @wh1te909.
Thanks for the response.

I've tried that and the error that comes back seems to be the exact same error that I was receiving in Postman.

After runnig the above api I looked through the logs for the environment and there were zero errors.
And just to make sure it wasn't the Agent ID being wrong, I tested the agent and successfully performed a reboot of the device via api.

Are there any debug conditions I can enable for the environment to get more details when attempting this api query?

Thanks.

[wh1te909]

500 errors are logged in django_debug.log search docs for this. Also I don't see you including any payload. You need to send a payload with the post request, like the id of the script, the timeout, script args, env vars etc. When I said to open browser dev tools and do the request, I meant from the tacticalrmm frontend, not from postman. Run an actual real script on the agent and then in dev tools you will see the exact json payload being sent.

[AshleyJackson]

Ah, gotcha.
Just wasn't seeing any details about what payload to send in the swagger docs.

I'll try testing the script payload now and get back to you.

Thanks in advance.

[AshleyJackson]

So that ended up working and I was able to run scripts via API.
Thank you for that.

The follow up issue I am witnessing is that the Gateway is timing out after 60 seconds, when the script timeout is set to 120 seconds.

Is there some configuration that I am missing to prevent the gateway from timing out like this?

Thanks for your time.

[wh1te909]

check nginx config

[wh1te909]

the nginx timeout set during install is 300 seconds unless you changed something. if you need longer than 300 seconds you should use the fire and forget method, there is no limit on the timeout for that.