Every time I boot and hit ESC to be able to see what is going on, I notice that as soon as the dkms checks begin, each dkms package check takes an excessive amount of time, ranging from 15 seconds for vboxadditions to 45 seconds for virtualbox. In all cases, the module for the kernel being booted has already been built and installed, and the oot message that finally come out say so. Any ideas about what would take so much elapsed time on a reasonably fast 8-core processor to determine that the module for the kernel is already installed ? Or where any additional dkms output not routed to the console might be ? Reproducible: Steps to Reproduce:
One possible contributing factor is that this is a regularly updated cauldron box that has about 32 older kernels in addition to the current one. But I still don't see why it would be necessary to process any of the older ones as I always boot the current one.
Keywords: (none) => TriagedCC: (none) => mageia, tmb
It's the 32 kernels. I just did a fresh reinstall on this system, and this no longer occurs when booting the new partition. But comment#1 still stands.
Still happening in current cauldron.
Assignee: bugsquad => kernel
Ping ?
No longer happening.
Resolution: (none) => FIXEDStatus: NEW => RESOLVED