Enable user activity logging

Last modified: 31 October 2022

User activity logging is a feature that allows you to log user activities. Below is an example of logged user actions: 

$ tail -f /tmp/datalore/datalore.log
{"@timestamp":"2022-10-11T14:46:14.594+02:00","@version":"1","message":"Audit logging event","logger_name":"audit","thread_name":"Computation edt_0_0","level":"TRACE","level_value":5000,"id":"B9otorSZuP2n2vyTHCkhLe","timestamp":"1665492374594","event_type":"run_cell","user_id":"pEoHIeNPj4DE43XInXC5OV","cell_content":"print(123)","file_id":"pEoHIeNPj4DE43XInXC5OV\/cEFzZVm3xAYFg3nvWeZqnT"}

Enable user activity logging

  1. Make a local copy of the logback.xml config file that can be found at: 

    services:
  datalore:
    ...
    volumes:
      ...
     - "./logback.xml:/opt/datalore/jetty/resources/logback.xml"

    .

  2. Edit the file copy. Choose a logback appender, add the net.logstash.logback.encoder.LogstashEncoder encoder for it and specify your logger configuration in place of <logger name="audit" level="OFF"/> as shown in the example below.

    <configuration>
    ...
    <appender name="AUDIT" class="ch.qos.logback.core.rolling.RollingFileAppender">
        <file>/tmp/datalore/datalore.log</file>
        <rollingPolicy class="ch.qos.logback.core.rolling.TimeBasedRollingPolicy">
            <fileNamePattern>datalore.%d{yyyy-MM-dd}.log</fileNamePattern>
            <maxHistory>30</maxHistory>
            <totalSizeCap>30GB</totalSizeCap>
        </rollingPolicy>
        <encoder class="net.logstash.logback.encoder.LogstashEncoder"/>
    </appender>
    <logger name="audit" level="ALL" additivity="false">
        <appender-ref ref="AUDIT"/>
    </logger>
    ...
</configuration>

  3. Replace the content of the original logback.xml file (same directory from which you run Docker Compose) with the content of the edited copy.

note

All logs are saved to /tmp/datalore/datalore.log. This file will be created inside the container. Make sure the folder with your logs is in a persistent volume so they are saved after the pod is restarted.