Skip to content

Commit

Permalink
log all the errors of dataframe
Browse files Browse the repository at this point in the history
  • Loading branch information
SandhraSokhal committed Feb 1, 2024
1 parent 9cf40b3 commit 02e44c0
Showing 1 changed file with 11 additions and 6 deletions.
17 changes: 11 additions & 6 deletions src/scripts/glue_jobs/glue_job.py
Original file line number Diff line number Diff line change
Expand Up @@ -94,9 +94,14 @@ def resolve_choice_and_write_output_frame(self, transformed_frame, partition_key
self.logger.info("Glue job finished.")

def log_errors(self, dynamic_frame):
self.logger.error("The total number of error count in dynamic_frame is {} ".format(dynamic_frame.stageErrorsCount()))
error_record = dynamic_frame.errorsAsDynamicFrame().toDF().head()
error_fields = error_record["error"]
for key in error_fields.asDict().keys():
self.logger.error("{} : {}".format(key, error_fields[key]))
raise Exception("Job failed with error : {}".format(error_fields["msg"]))
self.logger.error(
"The total number of error count in dynamic frame is {} ".format(dynamic_frame.stageErrorsCount()))
df = dynamic_frame.errorsAsDynamicFrame().toDF()
last_error_msg = "UNKNOWN"
for row in df.rdd.collect():
error_fields = row["error"]
for key in error_fields.asDict().keys():
self.logger.error("{} : {}".format(key, error_fields[key]))
if error_fields["msg"] is not None:
last_error_msg = error_fields["msg"]
raise Exception("Job failed with error : {}".format(last_error_msg))

0 comments on commit 02e44c0

Please sign in to comment.