[Nut-upsdev] [Nut-upsuser] Powerware 9155 with SNMP card

[Arjen de Korte]

[...]

>> All lights are green. In fact, I would like to get the opinion of other
>> users with similar systems. The easiest way to get that, is to modify
>> the development version. I'll commit this later today, with some small
>> modifications to make it conform to the NUT standards.
> fine with me :-)
> what points did i miss to conform to the nut standards?

Battery runtime (which is supposed to be seconds) and the values returned
to indicate the battery state for instance. No big deal. :-)

>> Yes. But that will probably take some time, as this requires a
>> considerable amount of development effort. Changing a few lines of code
>> takes a lot less effort than rewriting a whole driver. I guess it might
>> take at least a year or so, before we would have something of production
>> quality. I don't want to keep you waiting that long.
> So you just volunteered to implement the missing stuff for the snmp
> driver? :-P

I'll add the things mentioned before, which should take just a couple of
minutes.

> The small change you proposed before seems to work, though it takes quite
> a while (about 30secs) to get the status from WAIT to OL.

That's the side effect of moving status_init() and status_commit() to the
upsdrv_updateinfo() function. It will take 'pollfreq' (default 30 seconds
indeed) for the first update. You can put them probably somewhere else,
but then you have to look if this doesn't cause that only the first
"ups.status" defined is used. Given the fact that this only occurs once
and clients don't complain about this, I don't want to take the time to
see if this can be improved. It will once we bite the bullet and rewrite
this driver. Until then, we'll have to rough it out. If you don't mind the
additional network load, you could also reduce 'pollfreq' to something
like 10 seconds or so.

Best regards, Arjen
-- 
Eindhoven - The Netherlands
Key fingerprint - 66 4E 03 2C 9D B5 CB 9B  7A FE 7E C1 EE 88 BC 57