...
...
...
...
...
...
...
...
...
...
| Jira Legacy | ||
|---|---|---|
|
...
|
Investigate current behaviour with emulating of infrastructure slowness
...
If we would see errors in logs like RebalanceInProgressException, NoStackTraceThrowable: Timeout, CommitFailedException: Offset commit cannot be completed since the consumer is not part of an active group - we need to investigate these particular cases separately why records processed long time, or why consumers not hearbitted by Kafka broker, but Kafka retry mechanism works good.
Incorrect handling business exceptions
...
For testing I modified handler to process DI_PARSED_RECORDS_CHUNK_SAVED in mod-source-record-manager to throw new RuntimeException("jobExecutionId is invalid.");
For this case we logged exception, but also commited message as successful processed(Pic. 1) and progress bar was stack on the same place forever.
...
Each service should implement error's handler to send to Kafka DI_ERROR event if processing was with exception. In handler's implementation need to make sure, that DI_ERROR thrown for particular record, not for all file.
User stories:
MODSOURMAN-474 Implement ProcessRecordErrorHandler for Kafka Consumers - DRAFT
| Jira Legacy | |||
|---|---|---|---|
|
MODINV-408 Implement ProcessRecordErrorHandler for Kafka Consumers - DRAFT
MODINVOICE-252 Implement ProcessRecordErrorHandler for Kafka Consumers - DRAFT
MODSOURCE-290 Implement ProcessRecordErrorHandler for Kafka Consumers -DRAFT
|
| Jira Legacy | ||||||
|---|---|---|---|---|---|---|
|
| Jira Legacy | ||||||
|---|---|---|---|---|---|---|
|
| Jira Legacy | ||||||
|---|---|---|---|---|---|---|
|
| Jira Legacy | ||||||
|---|---|---|---|---|---|---|
|