An alarm is triggered if certain conditions are met. The value obtained is compared to the two values in one of four ways:
These are referred to as the Alarm Operator. The values compared to are referred to as 'Alarm Warn Value' and 'Alarm Critical Value'. Here is an example with the 'equal' operator:
Another example with the > indicator:
If the Host of an Alarm matches the FQDN or IP of a device (and is in the same Workspace) then the Alarm is said to be attached to the Device. If the Device has focus then the properties of the alarm target will be displayed along with the device properties. If the 'Side Bar' has 'Graph' selected, then a representation of the alarm state is displayed in a graph form. Beneath the graph representation, the up-time of the alarm will be shown. This is the % of time the alarm target spends in each state.
The amount of time displayed on the graph of an alarm target can be controlled by the Timeline. This appears above the alarm graphs in the side panel. The blue section of the Timeline represents the amount of time currently displayed. The toolbar above the Timeline allows for zooming in and out of the time period shown. The Timeline can also be shifted forward and backward with the toolbar. The dotted edge of the Timeline can be dragged so the graph shows more or less time. This enables you to select an interval and examine the state of the hosts in detail for the required period. The blue section can be dragged backwards and forwards. This moves the time shown in the alarm graphs backwards and forwards.
Any change in an alarm state is recorded in the logs.
Right clicking on a device with alarms attached will offer some useful functions. The first is to add an alarm to the host. You will not need to set the host if the wizard is started from this point. The option 'Check all Alarms Now' will ask the Aquirer for all alarms attached to the device to perform a test now. You will not need to wait for the scheduled next execution. The option to enable or disable all actions is also present. If a host is being brought down for maintenance and you do not want all the actions for the failures to be executed, disabling all actions could be very useful. Once the host is brought back on line you could enable the action and have all the targets 'Checked Now' to see if the host has started correctly. The option to reset all uptime counter's can also be found on this right click. You will also find the shortcuts to internal and external tools here, but more about those later.
If the workspace has focus and the 'Side Bar' is set to 'Graphs', a general health pie status graph for all hosts or services on the workspace will be shown. The 'Host Status' pie chart shows the number of hosts with all services OK, the number with some Critical and the number with some Warning. Hosts that have no checks are also represented. The tool bar allows you to select the 'Show Service Status' pie chart. This pie chart shows the number of services in each status as opposed to hosts with some services in a certain status.
Also See: Chapter 12, Logs
This value can be used to mask out a specific bit from the result obtained for the Target query. This can be very useful if the value obtained contains the status for more than one item. Lets say a value obtained shows the status of two different ports. The first port uses bit 1 to indicate all is OK, and second port uses the second bit. To get only the status of port one set the 'and Value With' 1, and for the second 2. You will need two different targets to test both.
Net-Probe performs a bitwise and on the resulting value of the target with the value specified. Setting the value to 0 disables the feature. Deleting the value will also disable it.
|Copyright (c) Warren Flemmer 2006||www.net-probe.com|