[Nut-upsdev] passing nut_debug_level from upsdrvctl to drivers

[Charles Lepple]

On Jan 27, 2008 4:35 AM, Arjen de Korte <nut+devel at de-korte.org> wrote:
>
>
> > Is there any reason why we couldn't pass the -D flags (counted by
> > nut_debug_level) from upsdrvctl to the individual drivers? It seems a
> > little strange to suggest to users that they should start the drivers
> > with upsdrvctl normally, but then start them by hand for debugging.
> > (If we want to debug upsdrvctl itself, we could use another flag for
> > that, instead of -D.)
>
> You're not the first to ask:
>
> http://lists.alioth.debian.org/pipermail/nut-upsuser/2007-May/002712.html

Sorry, it was late, and I didn't think to search first.

> I don't think this is a good idea. Passing a driver the -D option, will
> prevent it from backgrounding. When more than one driver is started by
> 'upsdrvctl' (you'd never know without looking at 'ups.conf'), it would
> result in an endless mess on the console. Therefor, we want to be sure
> that explicitly only one driver is started (and that we don't see
> interference from other drivers).

Ah, I didn't think of the case of more than one UPS.

> I also think we should not encourage people to post debug output before
> consultation with a developer. Most of the time, providing useful debug
> output requires more than just running the driver with an arbitrary number
> of -D flags anyway.

Well, the Tripp Lite units have had such buggy firmware in general
that I am tempted to just print out some debugging instructions in the
driver output, if we run across certain models.

-- 
- Charles Lepple