The Quality of Experience (QoE) is becoming an increasingly popular research area among the ICT related industry and academia.There is a need of a user-centric approach for the design and monitoring of network applications and services. It stimulates the need for an online passive mechanism to monitor the user activity in real-time. The mechanism may vary with the type of application. In this work, we keep our focus on Web browsing application as it is one of the most popular applications on the Internet. The aim is to let the users browse freely and monitor their activity instead of asking them subjectively about their usage experience. Monitoring of the TCP connection interruptions could be one of the ways to have indications about users’ feelings on the Web. This is based on the idea that, in case of bad performance, users break a Web browsing session by pressing a reset or stop button in the browser hence generating a TCP reset on TCP flow level. In this present work, we carry this analysis further to observe the user interruptions in relation to transfer sizes and durations. We want to see how the users react to bad performance and try to explain why we observe smaller mean transfer sizes. Do the users avoid launching large transfers in the case of bad performance or do they try to get the same files but stop the transfers that are becoming too long?