[Nut-upsdev] HCL: Zigor Danubio 2000 supported(ish) by blazer_usb

[Arnaud Quette]

2012/5/3 David Woodhouse <dwmw2 at infradead.org>:
> On Thu, 2012-05-03 at 20:45 +0200, Arnaud Quette wrote:
>> apart that load.* and shutdown.* may require the UPS to be on battery
>> to actually work.
>> at the very least, SxxRxxxx (S for shutdown, R for restart, and xx for
>> delays), so load.* and shutdown.*, must be implemented.
>> if this last test fails, you should really get refunded.
>
> Hm, wait — krauler_subcommand() doesn't even *do* anything for the
> SxxRxxxx commands. We're not actually asking the UPS to do anything at
> all.

this is part of the generic processing that is done in
blazer.c->blazer_instcmd()

> This test (test.battery.start.deep) should at least do *something*, but
> doesn't...
>
>  39.195653     send: TL
>  39.596730     received 10 (85)
>  39.596780     read: UPS No Ack
>  39.997623     received 10 (85)
>  39.997680     read: UPS No Ack
>  40.398541     received 10 (85)
>  40.398557     read: UPS No Ack
>  40.799400     received 10 (85)
>  40.799450     read: UPS No Ack
>  41.200526     received 10 (85)
>  41.200583     read: UPS No Ack
>  41.601580     received 10 (85)
>  41.601627     read: UPS No Ack
>  42.002598     received 10 (85)
>  42.002651     read: UPS No Ack
>  42.403712     received 10 (85)
>  42.403758     read: UPS No Ack
>  42.804772     received 10 (85)
>  42.804827     read: UPS No Ack
>  43.206129     received 10 (85)
>  43.206181     read: UPS No Ack
>  43.206195     instcmd: command [test.battery.start.deep] handled
>
> Did we ever have any documentation on this protocol, or was it all
> reverse-engineered?

Megatec/Q1 is a base protocol, which implementations vary depending on
the exact manufacturer and model.
there is no way to get a list of supported commands and settings, for example.

what we have is there:
http://old.networkupstools.org/protocols/megatec.html
http://old.networkupstools.org/protocols/sola.html

the rest is from users feedback on actual implementations.

it could be interesting to sniff the USB communication (with Wireshark
for example) using the Linux software, and do some testing around
commands (especially S.R...)

cheers,
Arnaud
-- 
Linux / Unix Expert R&D - Eaton - http://powerquality.eaton.com
Network UPS Tools (NUT) Project Leader - http://www.networkupstools.org/
Debian Developer - http://www.debian.org
Free Software Developer - http://arnaud.quette.free.fr/