[Nut-upsuser] OMNIVS1500XL and FreeBSD

[Charles Lepple]

On 3/3/07, Charles Sprickman <spork at bway.net> wrote:
> I'm running FreeBSD 6.2 with Nut 2.0.5 installed from ports...
>
> I have a TrippLite OmniVS1500XL plugged in, but I'm having little luck with the
> tripplite_usb module.  I have already built and booted a new kernel with UHID
> disabled so that libusb can grab /dev/ugen0.

Sounds good so far.

> At first I was able to get the following out of the driver:
>
> # USB_DEBUG=5 /usr/local/libexec/nut/tripplite_usb -u root -DDDDD auto
>
> Network UPS Tools - Tripp Lite OMNIVS and SMARTPRO driver 0.7 (2.0.5)
[...]
> send_cmd(msg_len=3, type='W')
> send_cmd: sending  3a 57 00 a8 0d 00 00 00 '.W......'
> usb_control_msg: 33 9 768 0 0xbfbfe960 8 4000
> send_cmd: received 57 00 0d 00 00 00 00 00 'W.......' (OK)

looks like the watchdog command works.

> send_cmd(msg_len=2, type='
> send_cmd: sending  3a 00 ff 0d 00 00 00 00 '........'
> usb_control_msg: 33 9 768 0 0xbfbfe960 8 4000
> usb_control_msg: 33 9 768 0 0xbfbfe960 8 4000
> usb_control_msg: 33 9 768 0 0xbfbfe960 8 4000
> send_cmd: received 00 20 01 58 58 58 58 0d '...XXXX.' (OK)
> send_cmd: send_try = 3, recv_try = 1

I'm a little concerned that it took three tries to send the command.

> Using OMNIVS 2001 protocol (2001)
> send_cmd(msg_len=2, type='S')
> send_cmd: sending  3a 53 ac 0d 00 00 00 00 '.S......'
> usb_control_msg: 33 9 768 0 0xbfbfe960 8 4000
> usb_control_msg: 33 9 768 0 0xbfbfe960 8 4000
> usb_control_msg: 33 9 768 0 0xbfbfe960 8 4000
> send_cmd: received 00 20 01 58 58 58 58 0d '...XXXX.' (bad)
> send_cmd: send_try = 3, recv_try = 3
>
> Could not retrieve status ... is this an OMNIVS model?
> [root at miko /home/spork]#
>
> Any ideas?

There are some #defines that you can tweak at the beginning of
drivers/tripplite_usb.c.

I would definitely not give up hope yet, though - I think it's just
some weirdness in the FreeBSD USB stack, interacting with the
empirical nature of this driver (the timeouts are just guesses based
on observed response times).

Also, someone submitted the patch for the 2001 protocol after some
experimentation, so I assume that they got status working at least
partially.

-- 
- Charles Lepple