Debugging errors

Preview error diagnostics

After a few minutes, you should notice that the Status shows a failure. Drill into the failure details to see what caused the failure.

  1. Click on Dataflow Run Start date
  2. Click on the Preview error diagnostics to see the specific details on each row that fail

Dataflow run status showing a failure

Preview error diagnostics link on the dataflow run details screen

The screen you now see shows you a bunch of details about what the error codes mean with the full error message and the row that failed.

Error diagnostics detail screen showing error codes, messages, and the failed row

NOTE
Scroll to the right side to see the source data associated with this error code

Understanding error types

INGEST-XXXX-XXX error

This error occurs because person.birthDayAndMonth is expected in the format of a two digit month plus a two digit day (i.e. April 27 should be formatted as 04-27)

The value (9-27) does not conform to the specified
regex pattern: [0-1][0-9]-[0-9][0-9] in field:
person.birthDayAndMonth of type: String
CAUTION
Note that person.birthDayAndMonth is not a required field but non-conformance to regular expression is treated by the system as a “data corruption problem” and is a serious error.

MAPPER-XXXX-XXX error

This error occurs because the source field of createDate has string values of Created on 2022-04-22T19:34:17Z. This value cannot be converted into a date automatically because of the text at the beginning: Created on. A calculated field must be used to cleanse the data.

Error transforming data for destination path
_dep.account.createDate. Details: Unable to convert
Created on 2023-09-24T10:19:58Z to schema type DATE_TIME
NOTE
This error is not a serious one as this leads to only warnings during mapping. The dataflow run does not fail because of this, so this lab does not fix this error.
recommendation-more-help
blueprints-learn-help-blueprints