Red Hat Bugzilla – Bug 204132
cpuspeed startup scrips fails in unduly noisy manner
Last modified: 2007-11-30 17:11:41 EST
Created attachment 134954 [details]
fix for a cpuspeed startup script
Description of problem:
/etc/init.d/cpuspeed is turned on by default but it may fail.
At least on a start it checks for that condition but only after
it produced a message which is not really relevant in the context
(cf. bug 201463 which, in a sense, may be not a bug at all).
'service cpuspeed stop' may just fail by trying to read from a
non-existent location and here a relevant check is just missing.
Sure, one can always do 'chkconfig cpuspeed off' but maybe problems
are only temporary specific to some kernels?
Attached is a proposed patch to deal with the problem.
Version-Release number of selected component (if applicable):
always on a machine where cpuspeed does not work
Eep, finally got mass-reassigned all the cpuspeed bugs over the weekend... I
believe this one is taken care of in the latest rawhide build. I like the
null-redirect on modprobing acpi-cpufreq, but not for $DRIVER. Since the user
must manually specify $DRIVER, I like spitting an error if they screwed up.