Nut upsdev - Jun 2007 - Belkin F6C1500-TW-RK won't stay powered off

Hello,

I'm using nut to control a Belkin F6C1500-TW-RK ups (vendor ID 050d, product
ID
0751).  I'm using the belkin driver and a serial connection for
communications.
 The host computer is running Debian.

I have things configured such that both my computer and the ups shut off when
the battery level reaches a critical level, and everything seems to work fine
up to this point.  My problem is that the ups doesn't stay off, even if the
line voltage hasn't been restored.  After being off for about 10 seconds,
the
ups turns back on, and subsequently supplies power to its outlets, which causes
my computer to restart.  After the computer boots and upsmon starts running, it
detects that the ups battery level is still critical, so the computer shuts
down again, along with the ups.  I haven't tested this extensively, but I
assume this cycle would continue until there is no longer enough battery power
to start the ups.

When the battery level reaches a critical level, I want the computer to shut
down and the ups to power off and remain off until the line power is restored.

Any suggestions as to why this is happening/how to fix it would be greatly
appreciated.

Matt Coffin

On 6/5/07, Matthew Coffin <MEWCOFFI at dal.ca>
wrote:
> Hello,
>
> I'm using nut to control a Belkin F6C1500-TW-RK ups (vendor ID 050d,
product ID
> 0751).  I'm using the belkin driver and a serial connection for
communications.
>  The host computer is running Debian.
Which version of NUT?
> Any suggestions as to why this is happening/how to fix it would be greatly
> appreciated.
Are there any system log messages from the driver indicating that
something is not working properly during shutdown?

Another thing you might try is to just put a dummy load (light bulbs,
etc.) on the UPS and run the shutdown command with debugging turned
on, e.g. "/lib/nut/belkin -k -DDD" (powering the system for the wall
during this test).

(Disclaimer: I don't know much about the Belkin protocol.)

-- 
- Charles Lepple

On 6/8/07, Matthew Coffin <MEWCOFFI at dal.ca>
wrote:
> Does this seem like a driver issue or a hardware issue?  With a computer,
the
> ability to boot up after line power has been restored is set in the
> BIOS, and I
> would have thought that the ups has similar low level logic built in, in
which
> case I should probably contact belkin to see if those settings are
> accessible. Would a driver be able to control such a low level
> behaviour?
[adding the list to the CC line]

It's entirely possible that your UPS uses slightly different commands
than the Belkin unit used to develop that driver.

This is the logic used to shut down in the belkin driver:

   http://boxster.ghz.cc/projects/nut/browser/trunk/drivers/belkin.c#L319

It does not look like there is any error checking (or even any return
code from the UPS that can be checked) so you may have to experiment.
Do the "load.on" and "load.off" instant commands work? That
might be
necessary for your UPS.

There are a number of options for figuring out the correct protocol
using the manufacturer's drivers on a supported OS.

-- 
- Charles Lepple