Managing execution errors

Execution errors can be categorized in a number of ways, but for the sake of simplicity we'll use a simple matrix to divide between design-time and run-time in one dimension, and anticipated versus unexpected in the other, to distinguish between the tools and approaches that are most likely to be used in each circumstance:

Handling anticipated errors

If we can anticipate that an error might occur, such as a database server being offline or unreachable, then we should be able to specify what should happen in the event of such an error. We should build our systems (ODI packages, scenarios, or load plans) to cope with this type of error.

The mechanism to use for this in ODI is the error (KO—or "not OK") path in the ODI Package editor. By using this technique we can send alerting e-mails, perform automated investigation or mitigation activities, and so on.

Tip

An important point to remember is that if our post-error processing completes successfully (for example, the alert...