Report types
There are five basic types of reports generated. They include:
- A verbose summary report of the activity during each time-slot (default).
- A terse version of the summary report (
--terse
option). - A down-time activity report (
--downtime
option). - A deny report (
--deny
option). - A grep-like copy of the access log records which match the search criteria (
--request
option).
Summary Report Categories
The verbose and terse summary reports concern a specific summary report category. These include:
- User-agent groups (
--groups
option). - User-agent sources (
--sources
option). - User-agent string (
--user-agents
or--agents
option). - URI (
--uris
or--urls
option). - Request address (
--ips
option). - Domain (
--domains
option, if domains are known). - HTTP method (
--methods
option). - Result-code (
--codes
option). - Referer (
--referers
option). - Request line (
--requests
option). - Authenticated user (
--users
option). - Protocol (
--protocols
option).
User-agents
For the purposes of classification, the user-agent strings are split into groups, and sources. These can be used to simplify the reporting when using the raw user-agent string is too complex.
User-agent groups
The User-agent group is a general classification of specific User-Agent strings based upon how the user-agent is normally used. The groups include:
- Bot – this group includes robots, spiders, crawlers and other automated access user-agents.
- Browser – A general browser, when more specific information about its environment is not available.
- Desktop – A browser used in desktop (and laptop) keyboard based operating environments.
- Mobile – A browser used in a non-keyboard operating environment.
- Tablet – A mobile browser known to be operating on a tablet.
- Phone – A mobile browser known to be operating on a phone.
- Game – A browser known to be operating on a game console or environment.
- Download – A tool (like wget), or library used to download web content.
- Proxy – A proxy interface between the actual requester and the site.
- Checker – A link or content checker.
- None – When no user-agent information is available. Either because it is not logged, or was not provided.
- Unclassified – When the user-agent is known, but has not yet been classified in a more specific group.
- Unwanted – this group includes site scrapers, spam email collectors and similar known user-agents.
- Unknown – Everything else.
User-agent sources
The User-Agent source is a generalization of the user-agent string based upon the owner or provider of the software. For browsers, this is the general browser-type: Chrome, Firefox, IE, Opera, et al. For bots, this is the controller of the service using the bot: Google, Yahoo, Baidu, et al.
Warning
The User-Agent database used to classify raw User-Agent strings into specific groups and sources is no longer being maintained. As time goes by, more and more User-Agent strings will be classified as Unknown for both the User-Agent group and User-Agent source.