[Nut-upsuser] bcmxcp_usb can not communicate with Eaton Powerware 5110

[Charles Lepple]

On Jul 13, 2012, at 12:41 AM, Greg Vickers wrote:

>   0.238682    get_answer: (128 bytes) => ab 01 79 01 02 50 00 50 01 00 0e 00 01 00 10 50
>   0.239445     4f 57 45 52 57 41 52 45 20 55 50 53 20 20 20 5c 00 00 00 00 00 00 00 00 00
>   0.240107     00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 51 51 00 00 00 00 51
>   0.240351     00 00 00 00 00 00 00 f0 00 f0 00 00 00 f0 00 00 00 00 00 00 00 00 f0 00 00
>   0.240558     00 00 00 00 00 00 51 00 00 51 00 00 00 00 00 00 00 00 00 f0 00 00 00 00 00
>   0.241334     00 00 00 00 00 00 00 f0 18 3b ab 01
>   0.241969    get_answer: block_number = 1
>   0.242121    get_answer: sequence number (1) is ok
>   0.242254    get_answer: checksum is ok
>   0.242891    get_answer: block_number = 1
>   0.243072    Communications with UPS lost: get_answer: not the right sequence received 0!!!

Well, that's odd.

I'm not an expert on this driver, but a quick read of that portion of the code implies that this shouldn't happen.

The last line says the received sequence number is 0. I think the sequence number is the 4th byte (my_buf[3]), which is 1 in the hex dump you posted:

http://trac.networkupstools.org/projects/nut/browser/trunk/drivers/bcmxcp_usb.c#L236

I was going to take a look at the source package to see if any extra patches were included, but it looks like there's a newer version of Debian for the Raspberry Pi (based on wheezy instead of squeeze), and it includes a newer build of NUT.

It doesn't look like much has changed in the NUT source tree between 2.4.3 and 2.6.4, but there's a significant chance that Debian wheezy was built with a newer compiler. Can you give that a try? If not, I'd recommend rebuilding NUT from source, so that we can patch in a few more debug statements.

Regards,

-- 
Charles Lepple
clepple at gmail