http://bugzilla.novell.com/show_bug.cgi?id=575647 http://bugzilla.novell.com/show_bug.cgi?id=575647#c1 --- Comment #1 from Miles Maybee 2010-02-01 15:57:17 UTC --- The problem is worse than first reported, as this small log excerpt shows: Feb 1 07:39:16 linux-j6pn pppd[25261]: Failed to open /dev/ttyACM1: No such device Feb 1 07:39:16 linux-j6pn pppd[25261]: Exit. Feb 1 07:39:32 linux-j6pn kernel: [58265.104078] usb 3-1: new full speed USB device using uhci_hcd and address 14 Feb 1 07:39:32 linux-j6pn kernel: [58265.293054] usb 3-1: New USB device found, idVendor=0fce, idProduct=d0d6 Feb 1 07:39:32 linux-j6pn kernel: [58265.293069] usb 3-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3 Feb 1 07:39:32 linux-j6pn kernel: [58265.293078] usb 3-1: Product: Sony Ericsson TM506 Feb 1 07:39:32 linux-j6pn kernel: [58265.293085] usb 3-1: Manufacturer: Sony Ericsson Feb 1 07:39:32 linux-j6pn kernel: [58265.293092] usb 3-1: SerialNumber: 3550760212173030 Feb 1 07:39:32 linux-j6pn kernel: [58265.293281] usb 3-1: configuration #3 chosen from 1 choice Feb 1 07:39:32 linux-j6pn kernel: [58265.322719] cdc_acm 3-1:3.1: ttyACM2: USB ACM device Feb 1 07:39:32 linux-j6pn kernel: [58265.329159] cdc_acm 3-1:3.3: ttyACM3: USB ACM device Feb 1 07:39:32 linux-j6pn kernel: [58265.339270] cdc_wdm 3-1:3.7: cdc-wdm-176: USB WDM device Feb 1 07:39:32 linux-j6pn kernel: [58265.346189] usb0: register 'cdc_ether' at usb-0000:00:1d.1-1, CDC Ethernet Device, 02:80:37:fd:02:00 Feb 1 07:39:32 linux-j6pn ifup: usb0 Feb 1 07:39:32 linux-j6pn ifup: No configuration found for usb0 On subsequent unplug/replugs, the device numbers continue incrementing... -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=575647 http://bugzilla.novell.com/show_bug.cgi?id=575647#c3 --- Comment #3 from Miles Maybee 2010-02-02 23:29:38 UTC --- While I'm connected, lsof | grep ACM shows, e.g.: pppd 23457 root 8u CHR 166,1 0t0 57621 /dev/ttyACM1 Upon unplug, lsof | grep ACM returns nothing. While connected, ls -l /dev/ttyACM* gives e.g.: crw-rw---- 1 root dialout 166, 0 2010-02-02 13:53 /dev/ttyACM0 crw-rw---- 1 root dialout 166, 1 2010-02-02 14:26 /dev/ttyACM1 crw-rw---- 1 root dialout 166, 2 2010-02-02 14:25 /dev/ttyACM2 upon unplug it shows: crw-rw---- 1 root dialout 166, 0 2010-02-02 13:53 /dev/ttyACM0 crw-rw---- 1 root dialout 166, 1 2010-02-02 14:26 /dev/ttyACM1 and so forth... a udev issue perhaps? When replugging, cdc-acm will not reuse an existing device node, it creates (or accesses) 2 new ones. Upon unplug, the higher number device node is deleted (even though it isn't the one being used), but the lower numbered node is not deleted, i.e., upon 1st unplug node 0 is not deleted, and cdc-acm uses nodes 1 and 2 upon replug. The second unplug leaves nodes 0 and 1, and 2nd replug generates nodes 2 and 3, etc ad infinitum. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=575647 https://bugzilla.novell.com/show_bug.cgi?id=575647#c5 Greg Kroah-Hartman changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |RESOLVED InfoProvider|miles.maybee@gmail.com | Resolution| |WONTFIX --- Comment #5 from Greg Kroah-Hartman 2010-08-25 23:14:07 UTC --- Yes, the userspace program is still hanging on to the device, and it needs to let it go before the kernel can reassign the same name again. So this is a userspace issue, nothing the kernel can do here. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.