[Nut-upsdev] Formalize use of the FSD flag (was: NUT driver update process...)

[Arnaud Quette]

2012/2/28 Arnaud Quette <aquette.dev at gmail.com>:
> Hi Charles
>
> 2012/2/28 Charles Lepple <clepple at gmail.com>:
>> On Feb 28, 2012, at 3:44 AM, Arnaud Quette wrote:
>>
>>> Hi Billy,
>>>
>>> switching back this thread to -upsdev
>>>
>>> 2012/2/27 William R. Elliot <bill at wreassoc.com>:
>>>> I have been able to set the 'FSD' from the driver and it is picked up by
>>>> upsmon.  That gives me the functionality that I was asked to put in the
>>>> driver.  If that is against the rules then I'll make an adjustment when the
>>>> allowed method is defined.  :-)
>>>
>>> yup, that's against the tables of the law ;-p
>>> more seriously, the right way to do this is to set the status to "OB
>>> LB", which will trigger upsmon to add FSD.
>>> the reason is that, in master / slave(s) setup, this allows upsmon to
>>> do the job, and notify slaves.
>>>
>>> so, as long as you just catch something that implies the FSD status
>>> using "OB LB", everything is fine.
>>>
>>> don't hesitate to post your patch on -upsdev to get feedback and
>>> adjustment request, which is needed before we can merge your contrib.
>>
>>
>> Arnaud,
>>
>> I looked at some of the existing NUT code for handling FSD, and I think it is a slightly more accurate representation of what is going on when an external source starts a shutdown on an UPS. See the following thread:
>>
>>   http://article.gmane.org/gmane.comp.monitoring.nut.devel/5882
>>
>> My thought is that the external shutdown signal may happen well before the battery is low.
>>
>> I don't see a reason why a driver couldn't set FSD, other than simply by convention.
>
> well, looking again at this, it's true that the notion behind FSD and
> OB+LB are not the same.
> moreover, we've been using FSD directly in some drivers (see
> netxml-ups, clone and clone-outlet for example) to notify an imminent
> shutdown.
>
> so you're right! In case a driver has a mapping on a data that is
> related to an imminent shutdown, FSD can be used.
>
> a patch is attached with a proposed modification to new-drivers.txt, §
> Status data.
> feedback and comment welcome.

I've just committed (r3478) a slightly modified version of this patch.

it should soon (~ 1h max iirc) be available to review:
http://trac.networkupstools.org/projects/nut/changeset/3478

cheers,
Arnaud