Web Server Access Log
  • 10 Jul 2024
  • Dark
  • PDF

Web Server Access Log

  • Dark
  • PDF

Article summary

This guide explains how to use a Web Server Access Log to see a commonly used log file which shows the calls made to the web server within Itential Automation Platform (IAP).

You will need to access the Profile application within Admin Essentials. Navigate to IAPAdmin EssentialsProfile. For more information, see Profiles.

Common Log Format

The log format for the Web Server Access Log is a stringified JSON object with each key mapping to a part of the Common Log Format. Since this is a JSON format, the ordering of the keys is not guaranteed. The keys, in the order they appear, in a traditional Common Log Format are referenced below.

Key Description
remote_addr The IP address where the request comes from.
remote_user The username within IAP that made the request. If the user is not authenticated, a value of anon will be used instead.
date The date the request was made in the Common Log Format.
method The HTTP method of the request (GET, PUT, POST, or DELETE).
url The url path used by the request.
http_version The HTTP version used to make the request.
status The status code returned as a response.
result_length The size of the response data in bytes. If this cannot be parsed, a value of "-" will be used instead.


The Web Server Access Log currently has two different configurable properties that can be edited within an IAP profile by updating loggerProps in the Profile Properties. Both properties can be found in the webserver section of the Configure tab. Example input is shown in the figure below.

Property Description
log_directory The file path to the directory where the access logfile will be stored.
log_filename The exact name for the web server access logfile that will be created and used to store all access logs. It will be located within the log_directory specified.

Figure 1: Profile Properties

Log Rotation

There is no automatic log rotation built into the Web Server Access Log which means that configuring an external log rotator is very important. One very useful tool to accomplish this is logrotate, a Linux tool that can automatically rotate a log file based on configured parameters. It can be installed simply by using your package manager and running:


yum install logrotate

Verify that logrotate is installed, and then check the configuration file, which for this example will be at /etc/logrotate.conf. There may be other ways to handle configuration such as inside the /etc/logrotate.d/ folder. The following is an example configuration where the webserver logfile is named webserver.log and is located at /var/log/pronghorn.

/var/log/pronghorn/webserver.log {
  rotate 10
  size 10M

In the above example, the logs will be rotated a maximum of 10 times before the old log files will be deleted. It will only rotate the logs when their size exceeds 10 megabytes, as specified by 10M in the configuration file. Finally, it will not compress files when it rotates them, as defined by nocompress, which it would otherwise do automaticallly if not specified.

Below is another possible configuration file:

/var/log/pronghorn/webserver.log {
  rotate 20
      echo Hello World

In this example, the logs will rotate weekly instead of rotating based on any fixed size. It will also rotate a maximum of 20 times. Finally, there is a postrotate script included, which means that in this case, any time the logs are rotated it will echo Hello World to the console. It may be useful to configure a custom script to run whenever logs rotate.

Running this command will start the log rotation:

    logrotate /etc/logrotate.conf

If you wish to force a rotation immediately, even when the various initiators of a rotation such as file size have not been met, run the following command:

    logrotate -f /etc/logrotate.conf

Querying Log Files

The log files can be queried using basic command line tools such as grep to search the file for a given pattern.

Using GREP or Similar Command Line Tools

If a user wanted to find any API requests made by a user named admin for the log file named webserver.log, they could use the following command:

grep '"remote_user":"admin"' webserver.log

In order to retrieve all the times a user has made a request to a specific route, such as GET /health/system for example, they could use the following command:

grep -E '"url":"/health/system"' webserver.log | grep '"verb":"GET"'

There are many other ways grep and similar command line tools can be used to query information from the JSON logs. Since all data values are preceded by a key, such as url or remote_user, it is possible to do standard queries on any value.

Using Third-Party Visualization Tools

The log format used by the access log should work with various third-party tools which support a standard JSON log format. These can be used for more advanced queries or visualization of logging information.

Was this article helpful?

What's Next
Changing your password will log you out immediately. Use the new password to log back in.
First name must have atleast 2 characters. Numbers and special characters are not allowed.
Last name must have atleast 1 characters. Numbers and special characters are not allowed.
Enter a valid email
Enter a valid password
Your profile has been successfully updated.