Nut upsuser - May 2014 - communication problems with eaton 9130 over usb

Hi!

I am not really experienced with UPS devices or UPS control software
like NUT but i am enthusiastic and trying to use Eaton-9130 (it has one
extra EBM too) as rationally as possible with NUT. I believe in general
things are under control and behave, although i have some glitches and i
would like to ask for help.

I am using Debian v. 7 amd64 and from its packages nut-server and
nut-client, v. 2.6.4-2.3+deb7u1. 

And these are my three problems communicating with UPS.

1. upsc show at different times different output using usb cable

# upsc eaton-9130
device.mfr: EATON Powerware 
device.model: 9130
device.serial: GG414A0357  
device.type: ups
driver.name: usbhid-ups
driver.parameter.pollfreq: 30
driver.parameter.pollinterval: 2
driver.parameter.port: auto
driver.version: 2.6.4
driver.version.data: MGE HID 1.31
driver.version.internal: 0.37
ups.mfr: EATON Powerware 
ups.model: 9130
ups.productid: ffff
ups.serial: GG414A0357  
ups.status: OB
ups.vendorid: 0463

If i restart /etc/init.d/nut-server stop and start, then other times it
gives, actually it is most frequent output

# upsc eaton-9130
battery.charge: 93
battery.charge.low: 20
battery.runtime: 64395
battery.type: PbAc
device.mfr: EATON Powerware 
device.model: 9130  3000VA-T
device.serial: GG414A0357  
device.type: ups
driver.name: usbhid-ups
driver.parameter.pollfreq: 30
driver.parameter.pollinterval: 2
driver.parameter.port: auto
driver.version: 2.6.4
driver.version.data: MGE HID 1.31
driver.version.internal: 0.37
input.frequency: 49.9
input.transfer.high: 276
input.transfer.low: 140
input.voltage: 231.0
input.voltage.nominal: 230
outlet.1.delay.shutdown: 7210
outlet.1.delay.start: 304
outlet.1.desc: PowerShare Outlet 1
outlet.1.id: 2
outlet.1.status: on
outlet.1.switchable: yes
outlet.2.delay.shutdown: 7210
outlet.2.delay.start: 301
outlet.2.desc: PowerShare Outlet 2
outlet.2.id: 3
outlet.2.status: on
outlet.2.switchable: yes
output.current: 0.40
output.frequency: 50.0
output.frequency.nominal: 50
output.voltage: 230.0
output.voltage.nominal: 230
ups.beeper.status: muted
ups.delay.shutdown: 20
ups.delay.start: 30
ups.load: 3
ups.load.high: 102
ups.mfr: EATON Powerware 
ups.model: 9130  3000VA-T
ups.power: 100
ups.power.nominal: 3000
ups.productid: ffff
ups.realpower: 58
ups.serial: GG414A0357  
ups.status: OL CHRG
ups.temperature: 23.9
ups.test.result: Aborted
ups.timer.shutdown: 0
ups.timer.start: 0
ups.vendorid: 0463

and still other times there exist some more entries, e.g. at least once
it had

ups.firmware: 0129

Is there some remedy for this, i.e. Debian v. 7 packages are known to be
broken and suggested way to resolve it is install manually from sources?
Or this 0129 ups firmware is known to be broken?

2. Second problem is that some times, though seldomly it says during
shutdown on serial console

...
Initiating UPS shutdown
Shutdown failed!
Driver failed to start (exit status=1)
[ ok down failed. Waiting for UPS batteries to run down.
[info] Will now halt.
...

3. upsrw sets somehow randomly outlet shutdown and start delay values,
it seems these are settable values i my case

# upsrw -u local_mon -p xxx eaton-9130 | egrep '^\['
[battery.charge.low]
[input.transfer.high]
[input.transfer.low]
[outlet.1.delay.shutdown]
[outlet.1.delay.start]
[outlet.1.desc]
[outlet.2.delay.shutdown]
[outlet.2.delay.start]
[outlet.2.desc]
[output.voltage.nominal]
[ups.delay.shutdown]
[ups.delay.start]
[ups.load.high]

And for example setting outlet.1.delay.shutdown gets changed some other
delay. I also have a feeling that these delays constitute some kind of
stack i.e. changing more that one time one value in row some other gets
changed. After some fiddling it is usually possible to get things done.
Practically it is not big problem since on ups panel i can change those
values too, although strange upsrw behaves like this.

What i intend to achive is really quite simple detection of LB and
shutting things down, and when power returns to UPS's mains, then start
things up again. But given the quite floating state of communication
with ups i have, i feel rather insecure, espesially considering this is
power control and almostly everything else depends on it.

Instead of usb cable i could use serial too, although i read that usb
cable is so to say mainstream way of communicating and serial is
considered legacy and is now around but maybe not for long. Do you
suggest using serial?

Also i have ups Web/SNMP Card available but i'd rather prefer get things
going with usb (or serial) because possible some other Eaton 9130
instances i probable will need to look after aint having Web/SNMP Card.
Or do you suggest intensively to use web/snmp card with nut when ever
possible as it is much more stable than usb?

Since i played around a lot with this UPS and Web/SNMP Card, during last
tests Web card is removed, maybe it is wise to do on ups control panel
'Restore Factory Settings'? (Or this way i destroy all calibration etc
what my reseller may have done and render ups into state it needs even
more love that it needs now?)

Sorry for a long letter and i would be thankful for suggestions!


Best regards,

Imre

BTW Saying on up control panel 'Start Battery Test' it answers test
could not be run at this time and try again later. Somehow a got to the
point where batteries got empty and started to charge again, now it is
92% and it says to try later again. Is this expected behavior and when
battery is 100% charged then battery test could be done?

On May 5, 2014, at 11:56 AM, Imre Oolberg wrote:
> Is there some remedy for this, i.e. Debian v. 7 packages are known to be
> broken and suggested way to resolve it is install manually from sources?
> Or this 0129 ups firmware is known to be broken?
Neither that firmware nor the Debian packages are known to be broken in that
regard. That said, you may want to check with Eaton to see if there is a
firmware upgrade.

Another potential problem is other packages that might be interrupting
communications with the UPS. Is this mainly a server, or are there other USB UPS
tools like upower running in a desktop session?

-- 
Charles Lepple
clepple at gmail