Analyzing Statistics for Sessions

The SESSIONS page of the Security Analytics dashboard allows you to analyze statistics relating to sessions carried out for different applications accessed via WSO2 IS.

For detailed information about the common functions of the Security Analytics dashboard, see Analyzing Statistics for Authentication Operations - Using the Security Analytics Dashboard.

Session Change Over Time

View (Example)

Description

This gadget indicates the following:

The number of currently active sessions.
The Active line indicates the number of active sessions over the selected time interval. A session needs to be active at the end of the selected time bucket in order to be counted as an active session.
The New line indicates the number of sessions that were started during the selected time interval.
The Terminated line indicates the number of terminated sessions over the given time interval.

WSO2 IS Analytics summarizes login and session statistics for every minute.

e.g., The table below illustrates the number of sessions terminated during the 21.19 - 21.20 time bucket.

Day	Hour	Minute	Active	New	Terminated
8	21	19	0	1	1

According to this table, a new session has started and then been terminated within the 21.19 - 21.20 time bucket. This new session is not counted as an active session for the 21.19 bucket because it was terminated within the same minute. The session was not active at the end of the 21.19 - 21.20 time bucket.

Purpose

This gadget allows you to:

Understand the load currently handled by your application in terms of the number of active sessions at any given time.
Understand the load handled by your application in terms of the number of sessions over any selected time interval.
Compare the load handled in terms of the number of sessions over different time intervals to identify patterns relating to the usage of your applications.

Recommended Action

Select different time intervals to identify the correlations between the usage of an application and time. When you identify the specific time intervals (e.g., specific times of the day, specific days of the week, etc.) when the usage of an application is particularly high, you can allocate more resources to handle the increased load. Likewise, you can allocate less resources during time intervals when the load is relatively low.

Top Longest Sessions

View (Example)
Description	This gadget ranks the longest sessions that have taken place during the selected time interval by the length of the session. The time duration for each session is displayed.
Purpose	This gadget allows you to: Identify the longest sessions that have taken place during different time intervals in order to identify any correlation between the session length and time. Identify users that carry out the longest sessions during different time intervals.
Recommended Action	Compare the top longest sessions for different time intervals. This allows you to identify the users who are most active during different time intervals.

Average Session Duration

View (Example)
Description	This gadget displays the average time duration of a session by each of the most frequent users for the selected time interval.
Purpose	This allows you to: Understand the average length of time spent by each user on a session. Observe changes in usage patterns for each user by viewing the average length of time for different time intervals.
Recommended Action	Compare the average time spent by each user in different time intervals to observe whether the session duration increases or decreases over time. Once this information is obtained, further investigations can be carried out to identify reasons for such user behavior and corrective action can be taken if applicable (e.g., increase/decrease in the efficiency of the application, an enhancement/depreciation of user experience which results in users spending more/less time on the application etc.).

Session Count

View (Example)
Description	This gadget groups sessions by their length and displays the number of sessions for each group during the selected time interval.
Purpose	This allows you to: Observe the count for different session groups and identify the general user behavior for your application in terms of the length of time spent on a session. Identify any changes in user behavior pattern in terms of the length of time spent on a session over different time intervals.
Recommended Action	Compare the session counts for different groups at different time intervals to observe changes in patterns relating to session length, and investigate further to understand the reasons for these changes (e.g., increase/decrease in the efficiency of the application, an enhancement/depreciation of user experience which results in users spending more/less time on the application etc.).

Data Table

View (Example)
Description	This gadget displays the complete list of sessions that took place during the selected time interval and provides detailed information about each individual session. The information provided about each session include the following: Username of the user who carried out the session The session start time The session end time (The time the session actually terminated, Due to logout event or forceful termination) The session termination time (The time the session is supposed to terminate) The duration of the session Whether the session is currently active or not The user store domain The IP address of the server The tenant domain Whether the remember me flag is set or not The time stamp The records in the data table sorted by the session ID by default, but they can also be sorted by other fields in the ascending/descending order if required as demonstrated above.
Purpose	This allows you to view detailed information about individual sessions.
Recommended Action	Sort the sessions displayed in the data table by different fields to identify the patterns of sessions managed by the application relating to Users, Session Duration, Start Time etc.