Computer Laboratory

Web site access logging

If you are curious about who is accessing your personal pages, you can analyse the logs of the main web server server. Note however that this will only tell you about direct accesses to our server; you cannot discover how many requests were satisfied from caches.

The Linux command cl-personal-www-log will extract the log entries relating to your personal pages. Command line arguments select which logs to look at, as follows:

The log file currently being written. This will usually contain log entries since the previous midnight. Owing to batching effects, entries made very recently may not be included.

The log file that was closed at the last midnight, which will usually contain yesterday's entries.

All closed logs currently held uncompressed - will usually be the last month or two.

All the old compressed logs. Note that this option is very expensive to run.

Includes logs of all pages instead of just your own, but with the calling host name anonymised for privacy reasons.

More than one of these options can be given if required. The log entries are written to stdout.

It is not guaranteed that the log entries will be delivered in monotonic time sequence, though they will be predominantly so. The "today" and "yesterday" options are not guaranteed to be implemented precisely.

All subsequent processing of the log entries must be done in accordance with relevant legislation, including the Data Protection Act, and our server's privacy policy.