the most wonderful thing about tiggers. Is tiggers are wonderful things.Why not use tiggers in binary kernel mod rpm's to at least attempt to get a more an up-to-date version? That way whenever the kernel changes the dependant rpm's would attempt to install the latest version of the modules.
However, triggers are not. They should be handled with excruciating care, like epochs. Now, in this situation you're describing you want an rpm to spawn off some other process to go fetch new kernel module rpms?
How would you standardize and mandate that? Remember, %scriptlets cannot be user-interactive.
-sv
