External Tests

An External Test is one that is run outside of Traverse (by a stand-alone script, for example). The test result is inserted into Traverse via the External Data Feed (EDF) and aggregated as though Traverse had collected it. Although the test itself is not run by Traverse, by creating an External Test, you determine how test results will be processed after they are received via EDF.

For more information on implementing external tests, see the Traverse Developer Guide & API Reference.

Creating an External Test

Navigate to Administration > Devices.
On the Manage Devices page, find the device for which you want to create a test and click Tests.
On the Manage Tests page, click Create New Advanced Tests.
On the Create Advanced Tests page, select the External Test option. Fill in the test name, test Interval, warning and critical thresholds, and, if desired, an action profile (optional). See the field descriptions in the table below.
Click Provision Tests.

Field	Description
Test Units	The units in which test results are displayed.
Maximum Value	Maximum possible return value for this test. You can generally ignore this unless you are using the test result to calculate a percentage of a whole. In that case, enter the value of the whole in this field. For example, if a test returns the number of MB available on a disk and you want to calculate the percentage of the disk's storage space that is available, enter the disk's total storage space in this field.
Result Multiplier	A number by which the test result is multiplied. If a test returns a number of bytes, for example, you can use a Result Multiplier of 8 to convert the result to bits.
Alarm After Inactivity	Number of minutes after which the DGE will mark stale test results as FAIL. The check is performed only if the DGE has received at least one test result since it was created. Use this to provide notification if no new results have been received for an external test.
Post Processing Directive	The computation applied to the test result after it has been multiplied by the Result Multiplier. Options include: Percent = current polled value / Maximum Value (e.g., current polled value represents 20% of total disk space). Delta = current polled value - last polled value (for example, 3 MB of disk space used since last poll). Delta Percent = (current polled value - last polled value) / Maximum Value (e.g., the difference between the current value and the last value represents 2% of total disk space). Rate = Delta / time between polls (e.g., rate of disk usage is 3 MB in 5 minutes). Rate Percent = percentage change since the last poll (e.g., rate of change measured as a percentage of the whole is 2% of total disk space in 5 minutes). Rate Invert = perform a rate calculation (2 consecutive poll, measure delta, divide by time) and then subtract the value from the configured maximum. Similar to Reverse Percent, but does not perform the % calculation. Reverse Percent = the difference between 100% and the percentage represented by the last polled value (e.g., last polled value for a disk usage test represents 20% of total disk space, so the reverse percent is 80%, which is the amount of free space). HexString to Long = poll an expected hexadecimal (base 16) value to convert it to base 10. For example the hexadecimal value 1A is converted to 26. Supports positive values only. TimeTicks = divide an expected timeticks value by 100 to convert it to seconds. None = polled value is not processed in any way.
As test value rises, severity:	Specify the relationship between test value and severity. Options include: Auto: If you select this option, Traverse sets this option based on the Warning and Critical thresholds for this test. If the Critical threshold is higher, as test value rises, severity ascends. If the Warning threshold is higher, as test value rises, severity descends. Ascends: As the value of the test result rises, severity rises. Descends: As the value of the test result rises, severity falls. Discrete: Specify a list of integers or ranges of numbers using the syntax: 1,3,5,10-25. Specify different values for warning and critical. Any returned value that does not match a value in either list means the device is OK. Bidirectional: You can set a "range" of numbers for each threshold and if the value crosses either of these two boundaries of the range, it will set the severity to Warning or Critical.