Nut upsuser - Dec 2024 - XCP documentation?

Once again, I am attempting to refine NUT support for my older 
(Eaton)Powerware Prestige 9 UPS units.

The issue is that the BCMXCP driver currently in NUT, optimistically 
assumes that everything in the XCP protocol document (C1 - as in the NUT 
archives) is available on all units, and that (even stated in the doc) 
is not the case.? For instance, the "Extended" block is stated to have
31 (max) bytes in the doc - my units offer 18 - and (not sure how/why 
yet) NUT tries to get values for unsupported values, and instead of 
seeing that it ran off the array, it is reading ??? and giving garbage 
values that corrupt some calculations and the plot ranges in 
upsimage.cgi. (And example are the max and min supported voltages . . . ).

So, while not asking for any help with that (since these are quite old), 
what I was wondering is if the NUT team managed to get a hold of any 
more of the old Exide/Eaton/Powerware XCP spec docs.? What is there 
mentions referring to the specific documentation for a given UPS for 
it's XCP implementation, and if that can be found, it would be a darn 
sight easier than trying to reverse engineer everything. (What I do get 
seems OK for the better part, but some values that *should* be supported 
seem to get nonsensical values, making me wonder what/if might be 
different in these units.).

Any pointers or docs are welcome!? (I've had to beat on this driver 
enough to get it working well, that I am good with "tweaking" it
myself
at this point . . . ).

Oh, and this is bcmxcp on serial . . .

Thanks,

- Tim

-- 
Tim Dawson

972-567-9360

Nothing?

Wow . . . 

- Tim

On December 4, 2024 3:03:41 PM EST, Tim Dawson <tadawson at tpcsvc.com>
wrote:
>Once again, I am attempting to refine NUT support for my older
(Eaton)Powerware Prestige 9 UPS units.
>
>The issue is that the BCMXCP driver currently in NUT, optimistically assumes
that everything in the XCP protocol document (C1 - as in the NUT archives) is
available on all units, and that (even stated in the doc) is not the case.? For
instance, the "Extended" block is stated to have 31 (max) bytes in the
doc - my units offer 18 - and (not sure how/why yet) NUT tries to get values for
unsupported values, and instead of seeing that it ran off the array, it is
reading ??? and giving garbage values that corrupt some calculations and the
plot ranges in upsimage.cgi. (And example are the max and min supported voltages
. . . ).
>
>So, while not asking for any help with that (since these are quite old),
what I was wondering is if the NUT team managed to get a hold of any more of the
old Exide/Eaton/Powerware XCP spec docs.? What is there mentions referring to
the specific documentation for a given UPS for it's XCP implementation, and
if that can be found, it would be a darn sight easier than trying to reverse
engineer everything. (What I do get seems OK for the better part, but some
values that *should* be supported seem to get nonsensical values, making me
wonder what/if might be different in these units.).
>
>Any pointers or docs are welcome!? (I've had to beat on this driver
enough to get it working well, that I am good with "tweaking" it
myself at this point . . . ).
>
>Oh, and this is bcmxcp on serial . . .
>
>Thanks,
>
>- Tim
>
>-- 
>Tim Dawson
>
>972-567-9360
>
>
>_______________________________________________
>Nut-upsuser mailing list
>Nut-upsuser at alioth-lists.debian.net
>https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/nut-upsuser
-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://alioth-lists.debian.net/pipermail/nut-upsuser/attachments/20241210/604062a4/attachment.htm>