options:
Insert a html Refresh tag into all index.html files. The value is
the refresh intervalle in minutes. Default to 5 minutes. Can be
overridden at command line with option -R | --refresh
To avoid loading index.html pages with incomplete content when SquidAnalyzer
is writing into the file, SquidAnalyzer first write into a index.html.tmp
file and once it is written it rename it into index.html.
Thanks to Weverton Everaldo Lubask and Pablo Ariel Tofano for the feature
request.
options:
Insert a html Refresh tag into all index.html files. The value is
the refresh intervalle in minutes. Default to 5 minutes. Can be
overridden at command line with option -R | --refresh
To avoid loading index.html pages with incomplete content when SquidAnalyzer
is writing into the file, SquidAnalyzer first write into a index.html.tmp
file and once it is written it rename it into index.html.
Thanks to Weverton Everaldo Lubask for the feature request.
follow:
Adjust timezone to use when SquidAnalyzer reports different time
than graphs in your browser. The value must follow format: +/-HH.
Default is to use local time. This must be considered as real
timezone but the number of hours to add remove from log timestamp to
adjust graphs times. For example:
TimeZone +01
will append one hour to all graphs time series to match squid server
timestamp.
Be warn that if you have enabled this directive before this commit the
graphs might display the wrong time. If your browser is at the same
timezone as the squid server you might want to comment this directive as
SquidAnalyzer will adjust the graphs to the right timezone by default.
Thanks to Weverton Everaldo Lubask for the report.
if an history file is present:
--skip-history : used to not take care of the history file. Log
parsing offset will start at 0 but old history
file will be preserved at end. Useful if you
want to parse and old log file.
--override-history : when skip-history is used the current history
file will be overriden by the offset of the
last log file parsed.
Thanks to zeft2016 for the feature request.
Of security reasons, store the squidanalyzer pages data (html, images...) data NOT in /var/www/html/squidanalyzer. Use insted /var/www/squidanalyzer for it.
Add feature to mention a different output directory.
Add feature to have CustomTitle just like CustomHeader.
Some bug fixes and optimization.
Ignore SIGHUP signal too.
a certain quantity in data file and sorted by OrderUrl. On huge access
log it will improve a lot the performances but you will have less
precision in the top url. Default to 0, all url will be stored.
Here are the performances of SquidAnalayzer on a 1.4 GB access log
file to parse and compute full reports over one week:
UrlReport | UserReport | Duration
----------+------------+---------
0 | 0 | 2m30s
0 | 1 | 3m00s
1 | 1 | 18m15s
1 | 1 | 9m55s when TopStorage is set to 100
specify a start and stop time. Log line out of this time range
will not be parsed. The format of the value is HH:MM. These
directives can be overriden with the -s | --start and -S | --stop
command line options. Thanks to Louis-Berthier Soulliere for the
feature request.
made in aliases files to avoid duplicates. You still have to use
--rebuild to recreate previous reports with new aliases. Enabling
this will imply a lost of performances with huges log files.
Gilles Darold
Gilles Darold
0xflotus
Klaus Tachtler
Amish