Ygtjki

On my personal laptop I'm using Fedora 29 and have two user accounts for myself; let's call them "personal_admin" and "personal_user". My personal_admin account is strictly for administrative tasks (installing software and updates mostly), which my personal_user account is for day-to-day use. Is this overkill?

How secure is switching between users using virtual terminals (CTRL+Alt+F#) versus the GNOME "Switch User" command? Does the situation change if I switch virtual terminals and then launch a GUI?

Suppose that my personal_user account is completely compromised and malware is running as a daemon with personal_user's privileges. I know that I shouldn't su personal_admin while signed into personal_user since the malicious daemon could easily record my personal_admin credentials. Does using virtual terminals or the GNOME "Switch User" command mitigate against this, or is the only safe option to sign out of personal_user (or maybe even reboot)?

From a security perspective, switching VTs directly and using GNOME’s “switch user” are equivalent. The “switch user” feature is more about making it user-friendly, than it is about security: it means you don’t need to know which VT you’re logged in on, or even whether you’re logged in yet.

If your two accounts are completely separate (in particular, neither can write anywhere the other account will read important information from), this will mitigate user-level compromise from one account to the other. Determining what constitutes a “safe” option for you really requires determining what risks you want to prevent, or minimise, and how far you’re willing to go to do so.