Hey guys,
The flow trigger that I configured for a job no longer seems to be working. When job 1 is complete, job 2 does not automatically start anymore. How can I fix this? Debugging is doubly difficult as the job seems to also be stuck in the queue.
Hey @Clem_Francis , looks like the changes you’ve made to that job are now returning 75,000 rows — rather than the 3k we tested. It’s probably crashing or getting killed by our state size limiter. Can we go back to the old query? (That was running end-to-end in about 4-5 seconds across all three jobs.)
In the meantime, I can look into why you’re not getting an alert for the aborted flow; sorry about that!
Hey @Clem_Francis I’m getting the below OpenFn error message in my email notifications every time the Job #1 runs. (FYI @taylordowns2000 - this type of error doesn’t show on the platform front-end, only in email notifications which is confusing for end users.)
Run 06435087-ef72-711d-96d0-b41bdd4e7844 of your project's timer job 5675
succeeded, but the resulting "final state" was 51.524MB and was too big to save
for subsequent executions.
Inspect job & take action: https://www.openfn.org/projects/os4h-dev-project/jobs/CGpWGj
Please check your job code and ensure that it's not growing during each run.
Common culprits are state.references and state.data, as if they are not manually
truncated some language-packages may allow them to grow.
Help altering your code to manage timer state size can be found in the OpenFn
documentation: https://docs.openfn.org/documentation/platform#managing-the-size-of-state-for-timer-jobs
Have you turned on your OpenFn Email Notifications? cc @rickettspc
To do so:
Each Time to turn on email notifications whenever there is a job failureThese notifications should help you understand what’s going on, but see Taylor’s question above regarding the query changes to consider what change you’ve made and why its now returning much more data than earlier.