Re: [Nut-upsdev] usbhid-ups on OmniOS (Solaris 10 derivate)

Sun, 21 Dec 2014 05:39:29 -0800

Hi,

here is more info:

uname -a
SunOS nas 5.11 omnios-10b9c79 i86pc i386 i86pc


# pkg info libusb
          Name: system/library/usb/libusb
       Summary: Sun wrapper library for libusb
   Description: Sun wrapper library for libusb (0.1.8)
      Category: System/Libraries
         State: Installed
     Publisher: openindiana.org
       Version: 0.5.11
 Build Release: 5.11
        Branch: 0.151.1.8
Packaging Date: Tue Mar  5 14:15:49 2013
          Size: 111.13 kB


I’ve also attach a debug info. Note how input voltage switches deom 240 -> 0 and back to 240 after a while (I’ve switched of AC), however “AC present” status stays always 1.

Matej

Attachment: apc_usb_on_batt_test.tar.gz
Description: GNU Zip compressed data

On 19 Dec 2014, at 14:23, Charles Lepple <[email protected]> wrote:

Arno,

what version of Solaris or libusb required the interrupt pipe hack?

- Charles

On Dec 19, 2014, at 7:00 AM, Arnaud Quette <[email protected]> wrote:

Hi Matej,

I've cc'ed the developers mailing list for now...

thanks for your report. Could you please send in some driver debug traces (starting it manually, using "usbhid-ups -DDDDD ...")?
That would help in troubleshooting the issue.

thx and cheers,
Arno

2014-12-19 8:48 GMT+01:00 Matej Sekoranja <[email protected]>:
Update: driver is not stable. I am switching to apcupsd-ups driver.

Matej

On 19 Dec 2014, at 01:21, Matej Sekoranja <[email protected]> wrote:

Hi,

I’ve recently installed OmniOS (running virtualized in VmWare ESXi with USB pass-through) and I got myself used APC SUA750I with USB port.
I’ve successfully installed NUT tools. All seemed to work fine…. however I got no change updates, e.g. if I disconnected AC there was no report about it via NUT.
Status was locked on OL, battery charge on 100% and so on… sometimes a value got updated, but this was very rare occasion (status was always OL).
System reboot helped to get recent values.

I’ve tried apcupsd and it worked flawlessly.
Then I wnt back to nut and I’ve noticed that interrupt processing is was disabled in NUT:

#if 1 //ndef SUN_LIBUSB
bool_t use_interrupt_pipe = TRUE;
#else
bool_t use_interrupt_pipe = FALSE;
#endif

A hack to a hack to enable it… fixed it. Now it works as it should.

It seems that SUN (well, Oracle) has fixed/updated their implementation… and requires IRQ to be processed.

Cheers,
Matej



_______________________________________________
Nut-upsdev mailing list
[email protected]
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/nut-upsdev



_______________________________________________
Nut-upsdev mailing list
[email protected]
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/nut-upsdev

Reply via email to