> > Is there a way to make the kernel boot in deterministic fashion
> > again? Everything will be done sequentially, and no task will kick
> > off if its predecessors are not done.
>
> I think I know the answer, but I'm going to ask anyway: why? I'm
> asking because there's probably more than one way to do it (There
> usually is in Linux.) and the best way will depend on what you need.
So I can be assured that variation in the boot process is eliminated as
a factor in any problem I'm seeing.
To use the NetworkManager case as an example, I boot with multi-thread,
see the error, save the boot record. Then, boot with single thread /
process, don't get the error, and save the boot record. Compare them,
and the clue to what is going wrong is highlighted.
Just FYI, if you're trying to deal with a networking dependency (e.g., you have multiple interfaces and you want to make sure they're all up before proceeding) that is broken by design and has been marked as "will not fix" by systemd, from what I can tell. System startup will proceed as soon as *any* network interfaces are online. If that's not your problem, then ... at least there's a chance there might be a fix for it.
-jdm
When I looked at the options, no obvious way to do this popped out, so
I thought I would tap the collective wisdom. If you know more than
one, I'd be happy to hear them all.
_______________________________________________
users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to users-leave@lists.fedoraproject.org
_______________________________________________ users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxxxxxxx
