This is an outdated version of the HTCondor Manual. You can find current documentation at http://htcondor.org/manual.

8.4 The CondorView Client Contrib Module

The CondorView Client contrib module is used to automatically generate World Wide Web pages to display usage statistics of a Condor pool. Included in the module is a shell script which invokes the condor_stats command to retrieve pool usage statistics from the CondorView server, and generate HTML pages from the results. Also included is a Java applet, which graphically visualizes Condor usage information. Users can interact with the applet to customize the visualization and to zoom in to a specific time frame. Figure 8.1 on page  is a screen shot of a web page created by CondorView. To get a further feel for what pages generated by CondorView look like, view the statistics for the University of Wisconsin-Madison pool by visiting the URL http://condor-view.cs.wisc.edu/condor-view-applet.

Figure 8.1: Screen shot of CondorView Client

After unpacking and installing the CondorView Client, a script named make_stats can be invoked to create HTML pages displaying Condor usage for the past hour, day, week, or month. By using the Unix cron facility to periodically execute make_stats, Condor pool usage statistics can be kept up to date automatically. This simple model allows the CondorView Client to be easily installed; no Web server CGI interface is needed.

8.4.1 Step-by-Step Installation of the CondorView Client

1. Make certain that the CondorView Server is configured. Section 3.12.6 describes configuration of the server. The server logs information on disk in order to provide a persistent, historical database of pool statistics. The CondorView Client makes queries over the network to this database. The condor_collector includes this database support. To activate the persistent database logging, add the following entries to the configuration file for the condor_collector chosen to act as the ViewServer.

       POOL_HISTORY_DIR = /full/path/to/directory/to/store/historical/data 
       KEEP_POOL_HISTORY = True
   

2. Create a directory where CondorView is to place the HTML files. This directory should be one published by a web server, so that HTML files which exist in this directory can be accessed using a web browser. This directory is referred to as the VIEWDIR directory.

3. Download the view_client contrib module. Follow links for contrib modules on the downloads page at http://www.cs.wisc.edu/condor/downloads-v2/download.pl.

4. Unpack or untar this contrib module into the directory VIEWDIR. This creates several files and subdirectories. Further unpack the jar file within the VIEWDIR directory with:

    
     jar -xf condorview.jar
   

5. Edit the make_stats script. At the beginning of the file are six parameters to customize. The parameters are

   ORGNAME

   A brief name that identifies an organization. An example is ``Univ of Wisconsin''. Do not use any slashes in the name or other special regular-expression characters. Avoid the characters $\mathtt{\backslash}$^ and $.

   CONDORADMIN

   The e-mail address of the Condor administrator at your site. This e-mail address will appear at the bottom of the web pages.

   VIEWDIR

   The full path name (not a relative path) to the VIEWDIR directory set by installation step 2. It is the directory that contains the make_stats script.

   STATSDIR

   The full path name of the directory which contains the condor_stats binary. The condor_stats program is included in the <release_dir>/bin directory. The value for STATSDIR is added to the PATH parameter by default.

   PATH

   A list of subdirectories, separated by colons, where the make_stats script can find the awk, bc, sed, date, and condor_stats programs. If perl is installed, the path should also include the directory where perl is installed. The following default works on most systems:

    
           PATH=/bin:/usr/bin:$STATSDIR:/usr/local/bin
   

6. To create all of the initial HTML files, run

           ./make_stats setup
   

   Open the file index.html to verify that things look good.

7. Add the make_stats program to cron. Running make_stats in step 6 created a cronentries file. This cronentries file is ready to be processed by the Unix crontab command. The crontab manual page contains details about the crontab command and the cron daemon. Look at the cronentries file; by default, it will run make_stats hour every 15 minutes, make_stats day once an hour, make_stats week twice per day, and make_stats month once per day. These are reasonable defaults. Add these commands to cron on any system that can access the VIEWDIR and STATSDIR directories, even on a system that does not have Condor installed. The commands do not need to run as root user; in fact, they should probably not run as root. These commands can run as any user that has read/write access to the VIEWDIR directory. The command

    
     crontab cronentries
   

   can set the crontab file; note that this command overwrites the current, existing crontab file with the entries from the file cronentries.

8. Point the web browser at the VIEWDIR directory to complete the installation.