Every n seconds check if slime repl is actively updated. if not, interrupt and call a restart that gives up on current operation that's causing the hangup e.g. sockets

slime-external-timeout.el

;; every 60 seconds check if slime repl buffer has been updated
;; if not intterupt and call restart-1
;; which should abort whatever is causing the hungup and continues

(defvar *slime-repl-last-output* "")
(defvar *slime-repl-watchdog-timer* nil)

(defun handle-sldb-restart ()
  (sldb-invoke-restart-1) ;; change here
  (message "Invoked restart-1 upon detecting a hungup."))

(setq *slime-repl-watchdog-timer*
      (run-at-time "30 sec" 30
              (lambda ()
                (with-current-buffer (get-buffer "*slime-repl sbcl*")
                  (goto-char (max-char))
                  (forward-line -1) ;;assuming the prompt has returned.
                  (let ((line (thing-at-point 'line t)))
                    (if (string= line *slime-repl-last-output*)
                        (progn (slime-interrupt)
                               (handle-sldb-restart))
                        (setq *slime-repl-last-output* line)))))))

(cancel-timer *slime-repl-watchdog-timer*)

(list-timers)