How the Worker is updating Execution status of the workflow

We are running n8n in queue mode, so once the worker pick the execution how it will update the execution is done, is it updating in the db or it again connecting with the main instance.

Hey @aashish.jain, I think that’d be a question for @krynble to say for certain. Based on the “Running a distributed system with this setup over SQLite is not recommended” hint in the documentation, my best guess is that workers would also write to the database (since concurrent writes aren’t great in SQLite), but Krynble would be able to confirm for certain.

you misunderstood the question.

Perhaps you could clarify your question then?

When the worker starts picking up the execution from the queue/redis and after execution how the worker is updating the successful execution of that workflow, is it directly updating to the DB or it is pushing to the queue again or is it connecting to main instance for updating the status

As mentioned before, based on the warning I suspect the worker would write directly to the database. But let’s wait for a definitive confirmation.

Sorry for the delay - we save all information to database regarding the completion and signal (via queue) the main process about the successful execution.

The main process can then respond or redirect this information if necessary to other sources, such as a webhook call.

No execution data (like payload) is transferred via queue to keep it lightweight.

Feel free to ask if you have any other questions!

1 Like