Instant mentoring services are novel social media, in which mentees can input expertise requests and wait for accepting some expert mentors who are willing to tackle the requests in an instant and one-by-one manner. While mentee's satisfaction of being mentored is determined by the matched mentor, this paper aims at analyzing and finding which mentors will respond to the given request raised by a mentee in Codementor, which is one of the popular instant mentoring services. We formulate the Mentor Willingness Ranking (MWR) problem. MWR is to understand whether a mentor is willing to tackle a request. We propose to deal with the task by generating a ranked list of mentors such that those mentors who are really willing to tackle the request are as many as possible. We develop three categories of features, Availability, Capability, and Activity, to model the willingness of a mentor dealing with the request. Results of analysis show the effectiveness of these features, and encourage develop learning-based method to accurately identify the willing mentors.