Nut upsuser - Feb 2014 - Eaton Powerware 5110 - some stats not reported

The driver will poll the UPS for what variables it supports when it starts, and
then go into a polling loop.  Look for a section defining the meter map, and you
should see the result, as well as the raw data.  You can get the comm spec for 
XCP from the nut site if you want to delve deeper. I'd give more specifics,
but am out of town and only working from memory at the moment . . .

- Tim

On February 14, 2014 11:22:22 PM CST, Greg Vickers <daehenoc at
iinet.net.au> wrote:
>On 13/02/14 12:16, Charles Lepple wrote:
>> On Feb 12, 2014, at 7:50 PM, Tim Dawson wrote:
>>
>>> Run the driver with -DDDDD and see what variables it picks up.  I
>have noted that with the Prestige 9 in bcmxcp, some variables are not
>consistent within models, and had to do some gentle remapping to get
>things to work correctly.
>
>I ran the new driver (see below) and with -DDDD it throws out tons of 
>never ending information, what am I looking for? :)
>
>> Tim, is this the issue you mentioned on 2014-02-02 with blocks 0x9
>and 0x13? I'll reply to that email.
>>
>>> On 02/12/2014 06:46 PM, Greg Vickers wrote:
>>>> Hi all,
>>>>
>>>> I've hooked up an Eaton Powerware 5110 to Ubuntu 13.10
server, and
>>>> configured nut appropriately.  The cgi scripts are working
fine,
>but on
>>>> upsstats.cgi, the Batter, UPS Temp and Battery Runtime fields
are
>all
>>>> blank - and there are no entries for these values in the
'All data'
>tree.
>> Greg,
>>
>>  From http://packages.ubuntu.com/saucy/nut-server it looks like you
>have 2.6.4. There was a big rewrite of the bcmxcp driver core which
>became part of 2.7.1.
>>
>> It might be possible to drop in the bcmxcp driver binary from Ubuntu
>trusty - I wouldn't force-install the whole package, just in case they
>changed things regarding init scripts and other libraries, but the
>drivers should be pretty self-contained. Or you could build just that
>driver from source, making sure to use the same paths as Ubuntu uses to
>build their binaries. Let me know if you're interested in trying one of
>those routes (admittedly, it makes for a bit of a Frankenstein system)
>and I can explain in more detail.
>
>Charles, thanks very much for the information :)  I've parachuted in
>the 
>bcmxcp_usb binary from Ubuntu Trusty from nut-server 
>(http://mirrors.kernel.org/ubuntu/pool/main/n/nut/nut-server_2.7.1-1ubuntu1_i386.deb)
>
>by using dpkg-deb to pull it out of the .deb file:
>
>$ ls -l /lib/nut/bcmxcp*
>-rwxr-xr-x 1 root root 75880 May 20  2013 /lib/nut/bcmxcp
>-rwxr-xr-x 1 root root 80544 Feb 15 09:03 /lib/nut/bcmxcp_usb
>-rwxr-xr-x 1 root root 71816 Feb 15 09:03 /lib/nut/bcmxcp_usb.2.6.4
>-rwxr-xr-x 1 root root 80544 Feb 15 09:02 /lib/nut/bcmxcp_usb.2.7.1
>
>And stopped both nut-client and nut-server, then started them both up. 
>
>There is no change to what is displayed in the cgi, the same fields are
>
>still missing :(
>
>On 13/02/14 15:55, Alf H?gemark wrote:
>> Hi
>>
>> On
>> http://nutwiki.kanonbra.com/wiki/Category:Eaton_Powerware_5110
>>
>> you can see what is reported. The unreleased part refers to v2.7.1,
>which is now released.
>
>OK, from that table, should the nut cgi be showing the Battery, UPS
>Temp 
>and Battery Runtime values?  Sorry, I don't know how these values are 
>translated into what is shown in the cgi displays.
>
>I've got nut running, and can use upsc to get the values directly:
>
>$ ps aux | grep nut
>nut      18306  1.3  0.0   2640   636 ?        Ss   15:14   0:00 
>/lib/nut/bcmxcp_usb -a Powerware5110
>nut      18308  0.0  0.0   2620   708 ?        Ss   15:14   0:00
>/sbin/upsd
>nut      18326  0.0  0.0   2592   764 ?        S    15:14   0:00 
>/sbin/upsmon
>
>$ sudo upsc Powerware5110
>battery.charge.low: 11
>battery.voltage:  13.5
>battery.voltage.low: 10
>device.mfr: Eaton
>device.model: POWERWARE UPS    700i
>device.part: 103004267-5591
>device.serial:
>device.type: ups
>driver.name: bcmxcp_usb
>driver.parameter.pollinterval: 2
>driver.parameter.port: auto
>driver.version: 2.7.1
>driver.version.internal: 0.28
>input.frequency:  49.9
>input.frequency.high: 55
>input.frequency.low: 45
>input.frequency.nominal: 50
>input.quality: 2
>input.transfer.boost.high: 216
>input.transfer.high: 280
>input.transfer.low: 186
>input.transfer.trim.low: 260
>input.voltage: 240
>input.voltage.nominal: 240
>output.current:   0.4
>output.current.nominal:   1.8
>output.frequency:  49.9
>output.frequency.nominal: 50
>output.phases: 1
>output.voltage: 240
>output.voltage.nominal: 240
>ups.beeper.status: enabled
>ups.description: Line-Interactive UPS, Single Phase
>ups.firmware: Cont:00.50 Inve:01.50
>ups.load: 24
>ups.mfr: Eaton
>ups.model: POWERWARE UPS    700i
>ups.power.nominal: 700
>ups.serial:
>ups.status: OL
>ups.test.result: Done and passed
>
>Happy to do more checks, I see the firmware on my 5110 700VA has the 
>same values as the 5110 500VA on the wiki, but there are a few fields 
>different, I guess there is a difference between the wiki page and the 
>driver in Ubuntu Trusty.
>
>Thanks,
>Greg
>
>_______________________________________________
>Nut-upsuser mailing list
>Nut-upsuser at lists.alioth.debian.org
>http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/nut-upsuser

---
Tim Dawson
Lewisville, TX

"Gaff tape can't fix "Stupid", but it can muffle the sound .
. ."
Sent via Android
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20140215/a5bbc064/attachment-0001.html>

> On February 14, 2014 11:22:22 PM CST, Greg Vickers <daehenoc at
iinet.net.au> wrote:
>> I ran the new driver (see below) and with -DDDD it throws out tons of 
>> never ending information, what am I looking for? :)
> 
On Feb 15, 2014, at 2:57 AM, Tim Dawson wrote:
> The driver will poll the UPS for what variables it supports when it starts,
and then go into a polling loop. Look for a section defining the meter map, and
you should see the result, as well as the raw data.
Tim: spot on.

Greg,

If you want, you can capture the -DDDD output to a file (until it starts
looping), gzip it, and post it to the list. I'm still getting up to speed on
the BCM/XCP protocol, but the meter map dump should have a few values we can
look up in the protocol spec.

Please excuse the mess; we're in the process of converting the protocol
library over to the same format as the rest of the website for ease of
maintenance:

http://old.networkupstools.org/protocols/eaton/
> On 13/02/14 15:55, Alf H?gemark wrote:
>> Hi
>> 
>> On
>> http://nutwiki.kanonbra.com/wiki/Category:Eaton_Powerware_5110
>> 
>> you can see what is reported. The unreleased part refers to v2.7.1,
which is now released.
> 
> OK, from that table, should the nut cgi be showing the Battery, UPS Temp
and Battery Runtime values? Sorry, I don't know how these values are
translated into what is shown in the cgi displays.
> 
From that table, yes, I think so. From the upsc output you posted, no. This
mismatch is odd, because driver.version.internal is 0.28 in both cases (this is
the version of the bcmxcp_usb driver, which generally gets incremented after
changes to that driver) and the firmware versions seem to match.

For reference, here's where the upsstats.cgi page gets its values:

https://github.com/networkupstools/nut/blob/master/conf/upsstats-single.html.sample

There is also a template for the table of multiple UPSes, but the mapping
between NUT variables and HTML table cells isn't as obvious.

-- 
Charles Lepple
clepple at gmail

On 16/02/14 00:38, Charles Lepple wrote:
> On Feb 15, 2014, at 2:57 AM, Tim Dawson wrote:
>> The driver will poll the UPS for what variables it supports when it
starts, and then go into a polling loop. Look for a section defining the meter
map, and you should see the result, as well as the raw data.
> Tim: spot on.
>
> Greg,
>
> If you want, you can capture the -DDDD output to a file (until it starts
looping), gzip it, and post it to the list. I'm still getting up to speed on
the BCM/XCP protocol, but the meter map dump should have a few values we can
look up in the protocol spec.
Thanks!  I've attached the file to this email 
(bcmxcp_usb.output2014160926.gz), it ran for 12 seconds, and looks like 
it's looping by that time.  I'm not sure what to look for, does grepping
for 'Yes' show the variables that the driver can pull off this UPS?
> Please excuse the mess; we're in the process of converting the protocol
library over to the same format as the rest of the website for ease of
maintenance:
>
> http://old.networkupstools.org/protocols/eaton/
Not a problem at all, I am happy to help with this driver :)
> On 13/02/14 15:55, Alf H?gemark wrote:
>>> Hi
>>>
>>> On
>>> http://nutwiki.kanonbra.com/wiki/Category:Eaton_Powerware_5110
>>>
>>> you can see what is reported. The unreleased part refers to v2.7.1,
which is now released.
>> OK, from that table, should the nut cgi be showing the Battery, UPS
Temp and Battery Runtime values? Sorry, I don't know how these values are
translated into what is shown in the cgi displays.
>>
>  From that table, yes, I think so. From the upsc output you posted, no.
This mismatch is odd, because driver.version.internal is 0.28 in both cases
(this is the version of the bcmxcp_usb driver, which generally gets incremented
after changes to that driver) and the firmware versions seem to match.
>
> For reference, here's where the upsstats.cgi page gets its values:
>
>
https://github.com/networkupstools/nut/blob/master/conf/upsstats-single.html.sample
>
> There is also a template for the table of multiple UPSes, but the mapping
between NUT variables and HTML table cells isn't as obvious.
Ah, I see - I've noticed that when I browse to upsset.cgi, that there 
are no Settings reported when I select Settings and click the View 
button.  If I select the Commands function and click View, there are 
only four available:
- Turn off the load and return when power is back
- Turn off the load and remain off
- Start a battery test
- Description unavailable

I have a Belkin UPS which I was running on this system up to a few days 
ago when I figured out that the battery was totally cactus, and I put 
the Eaton back in until I get replacement batteries for the Belkin.  
(The Belkin is two 12v batteries, the Eaton is one.) Anyway, the driver 
for the Belkin actually has Settings reported, and I can't remember, but 
about 10+ different Commands.

Is the lack of Settings or Commands indicative of any problem, or is it 
the case that these just haven't been implemented on this driver?

Cheers,
Greg

p.s. The only difference I can see from the web page any my upsc output 
is that my UPS is a 700, and the driver was developed on a 500?

-------------- next part --------------
A non-text attachment was scrubbed...
Name: bcmxcp_usb.output2014160926.gz
Type: application/x-gzip
Size: 5640 bytes
Desc: not available
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20140216/0724ff9e/attachment.bin>