Historically they've generally politely ignored community rebuilders and got a trifle enervated by commercial rebuilders - they changed how they handled distributing kernel code (to a fully patched tree rather than a pristine tree and a stack of patches I -think- from memory) in response to Oracle doing a commercial rebuild.
Exactly what the triggering incident was this time they've been very careful not to officially say (which is likely a better option than the optics of getting into a finger pointing war with a smaller target), and I suspect we won't be able to fully judge their motivations unless/until the details leak and/or are inferred by people close enough to the situation to guess correctly.
I don't know for sure but if I had to guess early Linux some 25-years ago didn't have the prevalence of polish it has now at least on the desktop. In the server space it was probably solid. That being said businesses back then were likely leery of running a RH clone with just some Linux staff -- better to pay RH for support.
Nowadays one could probably lean on staff to manage issues and arbitrage that go-it-alone mindset over paying the RH subscription. Now that loop-hole is closed.
Exactly what the triggering incident was this time they've been very careful not to officially say (which is likely a better option than the optics of getting into a finger pointing war with a smaller target), and I suspect we won't be able to fully judge their motivations unless/until the details leak and/or are inferred by people close enough to the situation to guess correctly.