Can you? YES
Should you? Well, depends on many factors, but quick answer is NO.
There are many articles written on the subject (a good start could be this one, which is very accurate yet funny and easy to read), but the important thing is to keep these tips in mind:
- be informative about what happened and what triggered the warning
- help the users know what to do to solve the issue
- try to solve the issue by yourself whenever possible
- lighten the mood and NEVER blame the user. Humor is a good idea to make error messages more friendly. And even though you might be convinced is the user's fault, it's always yours. Or at least that's what you should tell the user.
As for the degree of information needed, just provide AS MUCH AS NEEDED, AS LITTLE AS POSSIBLE. You can offer to expand info if needed, but most users will only need a brief. Take a look at Google's example below:

in this case, Google only shows the top part: Title and a brief description. However, you can see there's a lot more information when expanded (the image shows the expanded status). Thus: you have a brief description, but you offer the user to read more if needed. I think your case could be quite similar to this.
Warning vs User Action
Your mockup shows the possibility of acting upon the warning, which is something you can do, and it's done quite often. However, it's not a good idea, because you're educating users on being error prone.
If they can solve issues from warnings, not only you're mixing warnings and actions BUT you're teaching the user that if they make an error, they will be able to fix it once they do it. This is obviously wrong on many levels. On top of that, from a programming point of view, you'll need to have a system that works in a "normal" way, so to speak, and another that interacts with warnings, which is crazy and a lot of additional and completely unnecessary burden of work.
Thus, I'd recommend to tell your users the appropriate steps to follow from a proper area (for example, settings). Also, if needed, make sure roles and permissions are adequately defined. For example: are these certain actions on hardware based on permissions?
This way, by creating an area where users can choose this settings, you can centralize everything related to the behavior/interaction of hardware AND users can get back to this area in case of need. Otherwise.... they would need to make an error in order to get the option to solve it!