Comment # 6 on bug 917855 from Dr. Werner Fink

(In reply to Jan Engelhardt from comment #5)

You have been added to Carbon Copy list of this SLES bugs.


> >Revert-service-drop-support-for-SysV-scripts-for-the-early.patch
> 
> We do not have anymore boot.* scripts left besides boot.udev, and udev is
> pulled in through other ways already I presume.

Who ``we'' is?  If you're are talking about systemd upstream then be aware that
``we'' (which is openSUSE as well as SUSE) have to support this as this is how
old boot.* script had worked and have to work to be able to support SysVinit
boot scripts.  There are third party scripts which uses boot.* scripts as well.

> > module-load-handle-SUSE-etc-sysconfig-kernel-module-list.patch
> 
> There seem to be %post scripts in place - which create a
> /etc/modules-load.d/MODULES_LOADED_ON_BOOT.conf from the sysconfig parameter.
> That seems to make the patch superfluous.

Hmmm ... what does `seem to be' mean?

> >1087-infinit-timeout-for-kmod-loaded-modules.patch
> 
> There is absolutely no description to this. Judging from the leftover
> information ("infinit[e] timeout for kmod..."), I assume no good deed.
> Infinite timeouts are bad, they hold up the emergency shell forever.

This is what the kernel people had told us due to the fact that a killed worker
job loading a module is not the solution on large workstations or mainframes
where it could be that scanning and initializing hardware takes more time than
the hard coded timeout.  And ... which is the real problem ... killing this can
cause huge trouble.  Compare with bsc#889297