Main Components of Diagnosis

When you add a diagnosis to a NI, you can define various diagnosis actions to make the NI more flexible and verify the network design more accurately.

Table/Paragraph/Collector Variable

A variable can be a single variable such as $state or a table (paragraph, collector). For the table/paragraph/collector variable, you can add the Loop node for the system to loop through each table's row. Take the OSPF neighbor table as an example. There are four lines of neighbor information in the CLI result (the blue rows on the left). The diagnosis execution will determine whether the state contains full line by line (four lines in total).

You can also loop a list variable. See List Variable for more details.

Different Variable Types

Different types of variables have various operations such as Equals, Does not equal, Contains, etc. Compound variables and compound tables can be used in diagnosis definition as well.

Variable Type	Example
String	Note: For detailed description for the Compare Lines, see Appendix.
Int/Float

Different Data Source

For each variable defined in the diagnosis, you can select its data sources:
 

• Current: the value of this execution result of the NI .
• Baseline: the baseline data.
• Last: the NI will be executed twice and compare the current data with the last data.

The following diagnosis compares the current memory_usage with the baseline memory_usage:

Check Device Pair

You can compare the variables from the different devices. The following compares Mgmt_IP from two neighbor devices.

Use Compound Condition

You can have multiple simple conditions and combine them into a Boolean expression (and/or).
 

 Appendix: Diagnosis Operators