**** BEGIN LOGGING AT Mon May 21 02:59:59 2012 May 21 13:53:29 * e-jit has been updating for the first time since late 2009 May 21 13:55:05 I stayed with QtMoko for some time. As I was about to update, it occurred to me to try Android. After a few hours of not being impressed, I'm back to QtMoko (but now up-to-date) from OpenMobile.nl. May 21 13:56:02 After switching to Android (and this has remained the case on switching back to Qt Extended), I find the device no longer shows up as /dev/usb0 when plugged in via USB. May 21 13:56:27 This means I can no longer use the instructions at http://wiki.openmoko.org/wiki/USB_Networking#Simple_Manual_Linux_Configuration May 21 13:56:49 How can I set up the device and my PC such that I can ssh into the device ? May 21 13:58:42 This used to be easy ! May 21 14:06:04 didn't it appears as something else than usb0 ? try a dmesg on your linux host, eth2 by example ... May 21 14:06:22 ip a May 21 14:08:26 http://wiki.openmoko.org/wiki/USB_Networking#So_is_it_.22usb0.22_or_is_it_.22eth1.22.3F May 21 14:08:30 So is it "usb0" or is it "eth1"? May 21 14:08:43 The documentation below refers to the network interface name on the host side as "usb0". However, the actual name is determined by your host system ... May 21 14:09:24 ... As of the time of this writing, the Android and SHR distros will correctly pass the official mac address to the host, and will appear as "ethn" interfaces on the host. May 21 14:09:35 e-jit: did it help ? May 21 14:13:50 ... May 21 14:25:01 * e-jit didn't notice nullix's answer for a while - busy re-installing to make the most of QtBackup. May 21 14:26:21 I diff-ed the output of ls -1R /dev before vs after; it shows up in /dev/bus/usb/004/ and as a symlink in /dev/char/, but the ip address add $mask dev NAME command doesn't work with the full path or with char/189:402 in place of usb0, in commands that used to work. May 21 14:27:03 Meanwhile, /sbin/ifconfig reported no new interface. May 21 14:30:13 nothing displayed in dmesg when plugin the usb cable ? lsub ? May 21 14:30:40 I'll check dmsg in a minute, when I'm done re-installing. May 21 14:32:44 I got the following: May 21 14:32:53 [1010986.468019] usb 4-3: new full speed USB device using ohci_hcd and address 23 May 21 14:32:53 [1010986.645039] usb 4-3: New USB device found, idVendor=1457, idProduct=5122 May 21 14:32:53 [1010986.645043] usb 4-3: New USB device strings: Mfr=1, Product=2, SerialNumber=0 May 21 14:32:53 [1010986.645045] usb 4-3: Product: RNDIS/Ethernet Gadget May 21 14:32:53 [1010986.645047] usb 4-3: Manufacturer: Linux 2.6.29-rc3-v26 with s3c2410_udc May 21 14:32:53 [1010986.645141] usb 4-3: configuration #1 chosen from 2 choices May 21 14:32:54 [1010986.653681] usb0: register 'cdc_ether' at usb-0000:00:12.1-3, CDC Ethernet Device, 2e:88:ce:10:98:c4 May 21 14:33:02 but /dev/usb0 doesn't exist. May 21 14:34:32 i am not sure /dev/usb0 has to be created for cdc_ether kernel module May 21 14:34:44 ip a does not show any new eth ? May 21 14:35:16 * e-jit suspects "ip a" is an abbreviation, but doesn't know what for ... May 21 14:35:29 what exact command do I need to type to answer your question ? May 21 14:35:45 not at all 'ip a' is command ip with argument a... May 21 14:36:17 is usbnet loaded ? ( lsmod | grep usbnet ) May 21 14:36:45 Ah - OK. "ip a" reports eth1 and lo only. May 21 14:37:24 lsmod reports usbnet loaded (and in use by usbcore and mii) May 21 14:37:30 eth1 ? no eth0 ? May 21 14:39:03 d'oh - trying to concentrate on two things at once, I'd unplugged the device to restart Qt (it didn't pick up the network first time round) and forgot I'd done so when typing commands ! May 21 14:39:25 I now see usb0 in output of "ip a" - and yes, I get eth1 not eth0. May 21 14:40:14 ... and this time the ip address add works nicely, where it didn't before I unplugged the device ... May 21 14:40:22 Yay ! I have a usb connection ... May 21 14:41:21 remember that at some time after a QtMoko upgrade it might change from usb0 to ethX ... May 21 14:41:37 if i understand what the wiki states. May 21 14:43:08 i remember to have had this change, and some problem when freenrunner goes into sleep mode, it loose usb connection and does not restart it correctly at resume. May 21 14:43:55 Ah - OK. Thanks for hints and tips. May 21 14:44:29 * e-jit adds notes about "ip a" and usb0 -> eth? to script that remembers all this. May 21 18:08:20 hi **** ENDING LOGGING AT Tue May 22 02:59:58 2012