[systemd-devel] [RFC] [PATCHv3 2/3] resume: add a tool to write a device node's major:minor to /sys/power/resume.

[Lennart Poettering]

On Mon, 25.08.14 22:20, Ivan Shapovalov (intelfx100 at gmail.com) wrote:

> > > +        log_notice("Failed to resume from device '%s' (%s).",
> > > +                   device, major_minor);
> > 
> > same here..
> 
> Fixed as well.
> 
> > 
> > Hmm, what's the logic here? Is this something where we should halt the
> > machine? How "fatal" shall we consider this error?
> 
> > 
> > > +        return EXIT_SUCCESS;
> > 
> > ... especially, since you return "success" here...
> > 
> > I am just wondering whether log_notice() + EXIT_SUCCESS is the right
> > reaction here. Or maybe log_error() + freeze()? or maybe log_warning() +
> > EXIT_FAILURE?... Dunno. Just want to hear some rationale here... 
> 
> This is absolutely not fatal, and arguably not an error condition at all.
> 
> If a machine is simply rebooted, without hibernation image in place (but with
> resume= kernel parameter), this code path will be hit.

But then it should just be log_info(), not log_notice()?

Lennart

-- 
Lennart Poettering, Red Hat