Managing Systems and Groups

By default, there is only a single machine (the local system) listed in the Systems Table, which occupies the right part of the main window. You can add additional machines to monitor by pressing the Add System... button at the bottom of the window and then using the sheet (see Figure 1-2) that appears.

Using the menu at the left end of the sheet, you can choose to add one or more systems to the list in three different ways:

• Address— You can select individual machines connected via the Internet by typing in their IP addresses into the blank, either numerically or by name. Machines on the local network should generally be connected using their short names used for Apple file sharing (such as “dogboy.local”) instead of their full Internet names (such as “dogboy.mycompany.com”), as this method will work even if the local computers have temporary IP addresses acquired via a DHCP server. You may also choose a different IP port number using the text box to the right, if the default value of 47497 is blocked by your firewall.

• Range— You can select an entire range of numeric IP addresses to monitor. This is an excellent way to add an entire cluster of machines at once. As with the individual address option, you may also choose a different IP port number using the text box to the right, if the default value of 47497 is blocked by your firewall.

• File— You can select a text file containing a list of Internet addresses (name or number), one address per line. You can enter the name of the file manually in the text box, or press the “Select” button that appears at the right end of the sheet to This is a fast and effective way of adding a common set of computers. Alternate ports may be selected using the standard Internet notation of adding :<port number> to the end of each address.

Any machine that is added must have the /usr/libexec/statsCollector daemon running, as this daemon is responsible for actually doing the underlying statistics collection and shoveling data through the network to any copies of BigTop that are listening. BigTop will automatically start this daemon when the application is started, allowing both the local copy of BigTop and any remote systems to monitor that computer. You should note, however, that on systems where you cannot run BigTop — such as headless servers — you will need to start this daemon manually. The -p <port #> option should be used to set the port number if the default of 47497 is blocked.

The Systems Table alone is great if you just want to examine a few systems. For larger collections of computers, however, you may need more organization. The Groups Table, which occupies the left half of the main window, gives you a way to manage logical collections of machines on your network without continually adding and deleting those systems from the Systems Table. By default, BigTop shows you all of the machines being monitored (the “All” group). You can add groups to the Groups Table by selecting the Add Group button at the bottom of the window. Double-click on the new group in the Groups Table to edit the name. After creating a group, you can add machines to it by selecting the main (“All”) list and then dragging and dropping systems from the Systems Table to the group name. You can delete groups by selecting them in the Groups Table and selecting Edit→Delete (Delete).

The Systems Summary Table

The Systems Table lists all of the monitored machines (one per row) with a summary icon for each subsystem, allowing you to get a quick overview of the status of many machines at once:

• CPU— This gives an indication about the % of time that the processor(s) are active.

• Disk— This indicates the total number of bytes read from and written to disk per second.

• Kernel— The gives an indication of the number of kernel-managed threads currently executing on the system.

• Network— This indicates the total number of bytes coming in from and going out to the active network interfaces on this computer each second.

• Memory— This gives an indication about the % of the main memory that has been allocated.

• Processes— This gives an indication of the total number of processes.

Three additional types of statistics may be displayed or hidden by selecting the View→Show/Hide Advanced Statistics (Command-Shift-M):

• Load— This gives an indication of average system load (number of active, non-blocked threads) on the system.

• Swap— This indicates the % of available disk space that can be used for virtual memory page swapping has already been allocated.

• VM— This gives an indication of the number of page faults that are occurring per second on the system.

The availability of some kinds of advanced statistics depends on the version of the operating system running on the machine being monitored.

The state of each sub-system is summarized with an icon:

• Disconnected

• Low Utilization

• Medium Utilization

• High Utilization

The exact meaning of these various icons can be controlled using the “Alert Levels” editing function, as explained in Alert Levels Editor Window.