Nut upsuser - Jun 2013 - Cannot connect to a Eaton Ellipse Pulsar ASR/6000VA usbs [solved] but new problem surfaces.

Hi Charles, thank you for your answer.

This was indeed the core of the problem. I have amended the configuration
files.

It means now the communication works as required. Alas I now have another
problem.

Just for the record here is the configuration after the changes.

The configuration is as follows.

#nut.conf
MODE=standalone

#ups.conf
[eaton]
       driver = usbhid-ups
       port = auto


#upsd.users
[monuser]
password = password  #password changed to demo
upsmon master

#upsmon configuration
RUN_AS_USER monuser

MONITOR eaton at localhost 1 monuser password master #password changed to demo7

# --------------------------------------------------------------------------
# SHUTDOWNCMD "<command>"
SHUTDOWNCMD "/usr/lib/web-admin/backend.pl power_off"

Which solves the question from this post. The syslog messages do not show
errors.

Unfortunately when testing the UPS does not power down. The problem is the
upsdrvctl cannot pass the command to the ups.

this is the message from  upsdrvctl

louis at b3:/etc/nut$ sudo upsdrvctl shutdown eaton
Network UPS Tools - UPS driver controller 2.4.3
Network UPS Tools - Generic HID driver 0.34 (2.4.3)
USB communication driver 0.31
Using subdriver: MGE HID 1.18
libusb_get_report: error sending control message: Device or resource busy
Got disconnected by another driver: Device or resource busy
Can't initialize data from HID UPS
Driver failed to start (exit status=1)

I unplugged the usb and replugged and start nutted again, no change.

Hope you can help a little more, Thanks so far.

Regards, Louis

On 20 June 2013 01:53, Charles Lepple <clepple at gmail.com> wrote:
> On Jun 19, 2013, at 4:51 PM, Louis Chaillet wrote:
>
> > # Network UPS Tools: example ups.conf
> > [eaton]
> >        driver = usbhid-ups
> >          port = auto
> >       vendor = 0463
>
> Identifying the vendor or product is not strictly required, but if you
> provide 'vendor', it must match the 'Manufacturer:' value
('EATON', in your
> case).
>
> To match against the hex values, use 'vendorid' (or
'productid').
>
> Reference:
> http://www.networkupstools.org/docs/man/usbhid-ups.html#_extra_arguments
>
> --
> Charles Lepple
> clepple at gmail
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20130620/3ff88762/attachment.html>

On Jun 20, 2013, at 8:55 AM, Louis Chaillet wrote:
> Unfortunately when testing the UPS does not power down. The problem is the
upsdrvctl cannot pass the command to the ups.
> 
> this is the message from  upsdrvctl
> 
> louis at b3:/etc/nut$ sudo upsdrvctl shutdown eaton
> Network UPS Tools - UPS driver controller 2.4.3
> Network UPS Tools - Generic HID driver 0.34 (2.4.3)
> USB communication driver 0.31
> Using subdriver: MGE HID 1.18
> libusb_get_report: error sending control message: Device or resource busy
> Got disconnected by another driver: Device or resource busy
> Can't initialize data from HID UPS
> Driver failed to start (exit status=1)
> 
> I unplugged the usb and replugged and start nutted again, no change.
We could probably improve the documentation around this, but "upsdrvctl
shutdown" is meant to run late in the shutdown sequence, after the rest of
the NUT daemons (upsd and the drivers) are already stopped.

The reference to "upsdrvctl -t shutdown" is only meant to show the
shutdown order in a multi-UPS setup:

http://www.networkupstools.org/docs/user-manual.chunked/ar01s06.html#Shutdown_design

A usual end-to-end test would involve the "upsmon -c fsd" command,
which simulates a low battery condition. That does involve upsd, upsmon, and the
drivers.

Please let us know how we can clarify this part of the documentation.

-- 
Charles Lepple
clepple at gmail

On Jun 20, 2013, at 10:35 AM, Louis Chaillet wrote:
> With respect clarification of the documentation (and reference to your
question Charles). The documentation does suggest that you can shutdown the ups
by the upsdrvclt shutdown [ups] command. Specifically in
"http://www.networkupstools.org/docs/man/upsdrvctl.html" in the
section commands under warning. In the "Network UPS Tools User Manual"
(pdf) under section 6.3.2 in the last paragraph it is suggested to first test
the /usr/local/ups/bin/upsdrvctl -t shutdown and then "upsmon -c fsd".
Also this section suggests that under some circumstances you cannot rely on the
shutdown sequence to issue this command.
Thanks, noted.
> However the real reason I tried "upsdrctl -c shutdown" (not the
test) is that I  already tested the setup with the "upsmon -c fsd".
The results is that the server closes down, as expected, but ups does not
shutdown, and so there is no restart of the server.
> 
> I was guessing it has to do with the particular shutdown command of the
server I have, which was "SHUTDOWNCMD "/usr/lib/web-admin/backend.pl
power_off"". And I was trying to fix that by calling the upsdrvctl
shutdown command directly from a script.
> 
> As I do not use the packaged (debian) shutdown script (/sbin/shutdown) do I
need to tell the ups to power down myself, as   suggested in  the documentation
(configuration notes, see reference above)?
I haven't traced through the Debian shutdown sequence recently, but I think
that they call "upsdrvctl shutdown" after the rest of the system
processes have shut down (including the NUT driver). That should avoid the
"device or resource busy" error - only one process can open the UPS
USB device node at a time.

-- 
Charles Lepple

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20130620/05f97f88/attachment-0001.html>

Op 21 jun. 2013 05:02 schreef "Charles Lepple" <clepple at
gmail.com> het
volgende:
>
> On Jun 20, 2013, at 10:35 AM, Louis Chaillet wrote:
>
>> With respect clarification of the documentation (and reference to your
question Charles). The documentation does suggest that you can shutdown the
ups by the upsdrvclt shutdown [ups] command. Specifically in "
http://www.networkupstools.org/docs/man/upsdrvctl.html" in the section
commands under warning. In the "Network UPS Tools User Manual" (pdf)
under
section 6.3.2 in the last paragraph it is suggested to first test
the /usr/local/ups/bin/upsdrvctl -t shutdown and then "upsmon -c fsd".
Also
this section suggests that under some circumstances you cannot rely on the
shutdown sequence to issue this command.
>
>
> Thanks, noted.
>
>> However the real reason I tried "upsdrctl -c shutdown" (not
the test) is
that I  already tested the setup with the "upsmon -c fsd". The results
is
that the server closes down, as expected, but ups does not shutdown, and so
there is no restart of the server.
>>
>> I was guessing it has to do with the particular shutdown command of the
server I have, which was "SHUTDOWNCMD "/usr/lib/web-admin/backend.pl
power_off"".
And I was trying to fix that by calling the upsdrvctl shutdown command
directly from a script.
>>
>> As I do not use the packaged (debian) shutdown script (/sbin/shutdown)
do I need to tell the ups to power down myself, as   suggested in  the
documentation (configuration notes, see reference
above)?
>
>
> I haven't traced through the Debian shutdown sequence recently, but I
think that they call "upsdrvctl shutdown" after the rest of the system
processes have shut down (including the NUT driver). That should avoid the
"device or resource busy" error - only one process can open the UPS
USB
device node at a time.
>
Thanks for all your help, regards Louis
> --
> Charles Lepple
>
> A: Because it messes up the order in which people normally read text.
> Q: Why is top-posting such a bad thing?
> A: Top-posting.
> Q: What is the most annoying thing on usenet and in e-mail?
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20130621/81f73217/attachment.html>