**Important** This only happens on OpenRC Once usbmuxd was merged, I tried attaching iPhone to the computer. While the little phone icon has appeared at the control panel of KDE desktop environment, it brings me to a Dolphin page saying there was some error reading/connecting to the device. Further information (backtrace, emerge log, etc) will be provided soon. Possible workaround is https://github.com/libimobiledevice/usbmuxd/commit/e55e6e7c6cdcd7954de339be286b3818da69cddd, but 1) it isn't backported to 1.1.1, 2) I haven't got the chance to check whether it actually does anything. By the way, if this hanged running instance is killed by SIGKILL, and the device is reattached, everything works beautifully (without any issues) before the computer is restarted. Reproducible: Always Steps to Reproduce: 1. Launch PC 2. Attach Apple device Actual Results: Usbmuxd hangs Expected Results: Everything works as intended — the device shows in a file explorer and I can browse its files
Please, assign to me (I am the usbmuxd maintainer)
The issue has been (re)opened on the upstream bug tracker: https://github.com/libimobiledevice/usbmuxd/issues/210
FYI, usbmuxd doesn't start here with both original and the patched version. That could be it: https://github.com/libusb/libusb/issues/850 bConfigurationValue 0 <----- USB spec violation Here is the debug log: [00:33:05.442][3] usbmuxd v1.1.1 starting up [00:33:05.442][3] usbmuxd v1.1.1 starting up [00:33:05.442][4] Creating socket [00:33:05.443][3] Successfully dropped privileges to 'usbmux' [00:33:05.443][4] Initializing USB [00:33:05.443][3] Using libusb 1.0.26 [00:33:05.447][4] Registering for libusb hotplug events [00:33:05.447][4] Found new device with v/p 05ac:12ab at 1-46 [00:33:05.543][4] Setting configuration for device 1-46, from 1 to 4 [00:33:05.550][4] Found interface 1 with endpoints 04/85 for device 1-46 [00:33:05.551][4] Using wMaxPacketSize=512 for device 1-46 [00:33:05.551][4] USB Speed is 480 MBit/s for device 1-46 [00:33:05.551][4] 1 device detected [00:33:05.551][3] Initialization complete [00:33:05.442][3] usbmuxd v1.1.1 starting up As you can see, it's getting called 3 times, but doesn't stay in the memory for some reason. it works fine if I start it manually using the same command.
Created attachment 882351 [details] lsusb_info
Created attachment 882371 [details] udevadm monitor -p
btw, I'm openrc user, and the following PR may be required too: https://github.com/libimobiledevice/usbmuxd/issues/210
I assume you mean https://github.com/libimobiledevice/usbmuxd/pull/227?
I have rebased my patch onto 1.1.1 tag and published draft ebuild[1] (of -r2) with which issue seems to be gone. Could you test it? The overlay is here: https://gitlab.com/BalkanMadman/pidruchniki The patch is still a bit dirty, but that will be worked on. [1]: https://gitlab.com/BalkanMadman/pidruchniki/-/blob/master/app-pda/usbmuxd/usbmuxd-1.1.1-r2.ebuild
Please confirm whether this is still reproducible
So the upstream openrc patch is still not merged, see: https://github.com/libimobiledevice/usbmuxd/pull/227 https://github.com/libimobiledevice/usbmuxd/issues/210 What makes you to believe that the issue is fixed?
(In reply to Anton Bolshakov from comment #10) > What makes you to believe that the issue is fixed? If someone thought it was fixed, they would've closed it FIXED. They didn't; they need info and marked it NEEDINFO.
(In reply to Greg Kubaryk from comment #11) > If someone thought it was fixed, they would've closed it FIXED. They didn't; > they need info and marked it NEEDINFO. I was referring to "RESOLVED" status, which is still "closed". It was no changes at all, what's the reason to ask if is still reproducible? All info was provided, required patches highlighted. Users tested and published fixed ebuilds in their overlays.
(In reply to Anton Bolshakov from comment #12) > (In reply to Greg Kubaryk from comment #11) > > If someone thought it was fixed, they would've closed it FIXED. They didn't; > > they need info and marked it NEEDINFO. > > I was referring to "RESOLVED" status, which is still "closed". > It was no changes at all, what's the reason to ask if is still reproducible? > > All info was provided, required patches highlighted. Users tested and > published fixed ebuilds in their overlays. Sorry for not clarifying right away! Could you confirm whether the issue persists with =app-pda/usbmuxd-1.1.1_p20231011
