usbhid-ups regression (APC BE525-RS)
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
nut (Ubuntu) |
In Progress
|
Medium
|
Arnaud Quette |
Bug Description
Ubuntu 11.10
nut 2.6.1-2ubuntu2
UPS model: APC Back-UPS ES 525 (BE525-RS)
driver: usbhid-ups
I get the following error approximately every 30 seconds:
usb 4-2: usbfs: USBDEVFS_CONTROL failed cmd usbhid-ups rqt 161 rq 1 len 2 ret -75
The simalar error is described in the driver's known issues but increasing pollinterval doesn't help - the interval between error messages just increases accordingly.
I also noticed that battery.charge returned by upsc doesn't change when the UPS is actually discharging. I called upsc, unplugged UPS and called upsc again after a while: ups.status, battery.runtime and input.voltage reasonably changed but battery.charge stayed at 100%. So maybe the retreiving of this variable is working incorrectly.
Part of my ups.conf:
[be525]
driver = usbhid-ups
port = auto
pollinterval = 60
Output for "sudo /lib/nut/usbhid-ups -u nut -a be525 -DDDD" is attached.
I've tested a few other versions and got this:
2.4.3-1ubuntu3.1 - ok
2.6.0-1ubuntu3 - ok
2.6.1-2ubuntu2 - error
2.6.2-1ubuntu1 - error
2.6.3 (upstream) - error. Actually I didn't manage to run daemon that I built from source, so I ran just the driver and it reported the same error.
Looks like this problem appeared in v2.6.1.
Changed in nut (Ubuntu): | |
importance: | Undecided → Medium |
Hi Dmitry,
2012/1/13 Dmitry Andreychuk
> Public bug reported:
>
> Ubuntu 11.10
> nut 2.6.1-2ubuntu2
> UPS model: APC Back-UPS ES 525 (BE525-RS)
> driver: usbhid-ups
>
> I get the following error approximately every 30 seconds:
> usb 4-2: usbfs: USBDEVFS_CONTROL failed cmd usbhid-ups rqt 161 rq 1 len 2 ret -75
> The simalar error is described in the driver's known issues but increasing pollinterval doesn't help - the interval between error messages just increases accordingly.
> I also noticed that battery.charge returned by upsc doesn't change when the UPS is actually discharging. I called upsc, unplugged UPS and called upsc again after a while: ups.status, battery.runtime and input.voltage reasonably changed but battery.charge stayed at 100%. So maybe the retreiving of this variable is working incorrectly.
as per your provided log, both issues are related:
errno -75 is "EOVERFLOW" ("Value too large for defined data type").
We can see 1 successful retrieval of battery. charge (HID path: y.RemainingCapa city, ReportID: 0x0c), obtained through
UPS.PowerSummar
the interrupt pipe (ie, not requested explicitly, but sent by the
device itself, upon change)
All other request are failing, with EOVERFLOW, which should mean a
fault from the device (sending more data than it should
> I've tested a few other versions and got this:
> 2.4.3-1ubuntu3.1 - ok
> 2.6.0-1ubuntu3 - ok
> 2.6.1-2ubuntu2 - error
> 2.6.2-1ubuntu1 - error
> 2.6.3 (upstream) - error. Actually I didn't manage to run daemon that I built from source, so I ran just the driver and it reported the same error.
could you please send an output from a working version (preferably 2.6.0)?
cheers, powerquality. eaton.com www.networkupst ools.org/ www.debian. org arnaud. quette. free.fr/
Arnaud
--
Linux / Unix Expert R&D - Eaton - http://
Network UPS Tools (NUT) Project Leader - http://
Debian Developer - http://
Free Software Developer - http://