WLWatchman Manual :: First Monitoring Session |
Preliminaries | ||
If you've used earlier versions or even read the documentation, you'll know that WLWatchman has two modes of operation: command-line or console. Here we deal primarily with the console, though some words on the command line are here
Monitoring from the console | ||
Console mode is the default, or you can run startWLWatchman.sh with --console to run the GUI console. It will look for a file called wlwatchman-config.xml in the current directory. (Note, all other command line switches are ignored when running with the console, as it provides graphical means of setting them. If you don't like this, send me a patch!)
Monitoring from the command-line | ||
The command-line mode is intended to be used from a non-graphical terminal - many test/production environments do not provide graphical access and the only way to get at the box is a secure login session. In addition, some will block RMI access to the servers so your only choice is to run WLWatchman from the command line.
Start WLWatchman with --no-console to run in command-line mode. A full set of command-line switches are given here, though the ones of primary interest, assuming you have a wlwatchman-config.xml to use are --refresh and --repeat which determine how frequently and for how long the target domain is queried for data..
If you don't have a wlwatchman-config.xml already created then you'll want to run the first time with the --all switch, which will cause WLWatchman to fetch the list of available MBeans and attributes from all the (running) servers in the domain. Note that this list is potentially very large indeed, so you'll probably want to temper it with judicious use of --include-file or --exclude-file depending on whether you know the specific MBean types you want or don't want.
Note that WLWatchman defaults to looking for its config file as wlwatchman-config.xml in the current directory; however you can ask it to load any file using the --config-file switch.