Because no upgrade goes unpunished, Mu4e was refusing to start up in Emacs, throwing an “error in process sentinel: Mu server process ended with exit 1” error. I couldn’t find this exact error message online, but a look through posts about similar error messages revealed that complaints about the Mu server were often caused by the server still running. I took the advice in this Mu4e issue thread and ran:
pkill -2 -u $UID mu
…exactly like that, without replacing the $UID with anything.
Mu4e sprang right back up afterward, having grabbed control of the database back from the server process that had been running.
Still waiting wearily for more troubles to happen, but in the meantime at least I can read emails again.