You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Have you used the BPMN-Linter in Flow Designer to Validate your Diagram?
Yes
What happened?
Hello,
We made a mistake in the set timer definition on the first user task in a process.
The timer definition type was not correct, it should have been an interval. This caused an exception in
ORA-01841: (full) year must be between -4713 and +9999, and not be 0
ORA-06512: at "FLOWS.FLOW_TIMERS_PKG", line 473
Normally this exception is logged through the exception handling after the "case when" in the timer package, however, the logger has a design flow (flow_logging):
The log itself is written using an insert into select, where the select assumes that the flow_processes record with the p_process_id exists. In our case it did not, I guess because the exception is in the first step of the process and the insert into flow_processes would have come later. So the real problem was not logged and I came down the rabbit hole to understand what was happening 😊
Any kind of error logger should write the log no matter what, I suggest to do the select and the logging separately.
Steps to reproduce
see above
Expected behaviour
Log with the original exception
The text was updated successfully, but these errors were encountered:
Flows for APEX version
23.1 (latest)
Oracle Database version
19c
Oracle APEX version
23.1
Have you used the BPMN-Linter in Flow Designer to Validate your Diagram?
Yes
What happened?
Hello,
We made a mistake in the set timer definition on the first user task in a process.
The timer definition type was not correct, it should have been an interval. This caused an exception in
ORA-01841: (full) year must be between -4713 and +9999, and not be 0
ORA-06512: at "FLOWS.FLOW_TIMERS_PKG", line 473
Normally this exception is logged through the exception handling after the "case when" in the timer package, however, the logger has a design flow (flow_logging):
The log itself is written using an insert into select, where the select assumes that the flow_processes record with the p_process_id exists. In our case it did not, I guess because the exception is in the first step of the process and the insert into flow_processes would have come later. So the real problem was not logged and I came down the rabbit hole to understand what was happening 😊
Any kind of error logger should write the log no matter what, I suggest to do the select and the logging separately.
Steps to reproduce
see above
Expected behaviour
Log with the original exception
The text was updated successfully, but these errors were encountered: