We see the commit that caused this degradation since 1.10, you are correct that this bug exists for few weeks now. We will investigate internally how to improve the process for finding and resolving similar issues in the future.
I can assume that the reason we did not encounter it sooner is because it only impacts new gateways and not the existing configuration.
This is actually a very successful feature with high adoption rates, since introduced earlier this year, thousands of gateways have already configured it.
Not sure that you are familiar with it, but there is a feedback button at the top toolbar that will allow you to share feedback directly with RnD and report such issues, or share your general feedback on the application.
On another topic, I invite you to participate in R81.10.10 EA program, as it contains many new enhancements for Spark Management (SMP). Fill this survey to join if you have not done it yet: EA survey
Thanks again for your feedback.