On Aug 20, 2018, at 7:40 AM, Valentin Merkulov <[email protected]> wrote: > > Attached are config.log and usbhid-ups output with -u root (gzipped). > Now driver output looks a lot like the one from port build.
If memory serves, this was the sort of error we were seeing from 3016 devices on Linux: 0.154899 [D1] Path: UPS.PowerSummary.iManufacturer, Type: Feature, ReportID: 0x2b, Offset: 0, Size: 8, Value: 3 0.167518 nut_libusb_get_report: No device. 0.167531 [D1] Can't retrieve Report 30 (No device). After you see the "No device" error, does the UPS reattach under a new /dev node? Do the usbconfig commands return any data? > On Mon, Aug 20, 2018 at 1:42 PM, Charles Lepple <[email protected]> wrote: >> On Aug 20, 2018, at 1:03 AM, Valentin Merkulov <[email protected]> wrote: >>> >>> 0.033412 [D3] nut_usb_claim_interface: failed to claim USB >>> device (Other error). >>> 0.033417 [D3] nut_usb_claim_interface: failed to detach kernel >>> driver from USB device (Other error). >> >> I am not as familiar with this branch (I mostly worked with the older >> libusb-1.0+0.1 branch mentioned in issue #300), but this might not be using >> the libusb-1.0 API. I thought that the libusb-1.0-compat branch would pick >> up the 1.0 API in preference to the 0.1 API. Do you have the build logs? >> config.log would work as well, though it is relatively large. >> >> Another thing to try is to pass "-u root" to the driver, since we have had >> trouble distinguishing permission errors from other USB errors on some >> platforms. But I realize that you were able to get things working initially, >> so the devd files are probably still set up. >> >> (A lot of my spare UPS equipment is packed up at the moment, otherwise I >> would try this branch out on a FreeBSD system myself.) > <config.log.gz><usbhid-ups.output.gz> _______________________________________________ Nut-upsuser mailing list [email protected] https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/nut-upsuser
