not messianic enough - I forgot to check that the inverter was incommunicado. this is a different computer (t43 like the xp box) with different fedora (16 vs 15). I wasn't around that day. ok. so now I need to redo on a day that I know I'm around to confirm that the inverter gets nobbled. I'll confirm that the tusb interference is happening, too. Tim On 21 Jan 2012, at 20:40, Alan Stern wrote: > On Sat, 21 Jan 2012, Tim Coote wrote: > >> Ok. It looks to me like I hadn't got usbfs_snoop working for that last extract from the syslog. >> >> So, I built this script to set everything up: >> >> I rebooted the computer and ran the script, after sunset with the inverter connected. >> >> I then let the sun come up, > > Sounds a little messianic, there. :-) > >> stopped the wiresharks and saved their captured files. copied /var/log/messages and stripped out the noise with: >> sed '1,/Jan 16/d' messages |sed '1,/0\.000000/d' |egrep -v "NetworkManager|dhclient">filteredSyslog >> >> this leaves me with the following files in the current directory: >> >> -rw-------. 1 root root 16572 Jan 15 15:12 bus1PlugInPviForgotToRunUsbSnoop >> -rw-------. 1 root root 701780 Jan 17 07:50 bus1UsbPluggedInOverSunup >> -rw-------. 1 root root 4278 Jan 15 15:08 bus2PlugInPviForgotToRunUsbSnoop >> -rw-------. 1 root root 282 Jan 16 21:12 bus2UsbPluggedInOverSunup >> -rw-------. 1 root root 8495 Jan 15 15:13 bus3PlugInPviForgotToRunUsbSnoop >> -rw-------. 1 root root 13083260 Jan 17 07:50 bus3UsbPluggedInOverSunup >> -rw-------. 1 root root 1614 Jan 15 15:08 bus4PlugInPviForgotToRunUsbSnoop >> -rw-------. 1 root root 460 Jan 16 21:12 bus4UsbPluggedInOverSunup >> -rw-------. 1 root root 282 Jan 15 15:08 bus5PlugInPviForgotToRunUsbSnoop >> -rw-------. 1 root root 816 Jan 16 21:12 bus5UsbPluggedInOverSunup >> -rw-r--r--. 1 root root 628207 Jan 21 15:37 filteredSyslog >> -rw-------. 1 root root 2924156 Jan 17 21:02 messages >> >> Which, I think is the output that's needed. I think that the inverter >> must be plugged into bus3. > > That's right. > >> So I'm dropping bus3UsbPluggedInOverSunup >> and filteredSyslog into my public folder again. >> (http://dl.dropbox.com/u/10589123/filteredSyslog and >> http://dl.dropbox.com/u/10589123/bus3UsbPluggedInOverSunup >> >> When I look through filteredSyslog, I cannot see what other program's >> driving the device. How can I find what's happening as /dev/ttyUSB0 >> appears. Should I be running lsof or similar to get a view on what's >> waiting for the usb device to come up? I'm guessing that whatever it >> is is waiting on some file to appear/change or on a socket, but I >> don't know how to validate that, nor how to identify the program. > > This time the trace does not show any other programs interfering with > the TUSB device. Instead, the device starts up at 7:33:33 but doesn't > have sufficient power to run properly until 7:50:31. After that there > is no problem until 11:51:43 -- maybe a cloud covered up the Sun > briefly. Anyway, at that time the device disconnected but then > reconnected less than a second later. At 16:33:21 there was another > disconnect, followed by a few failed reconnections. And that's all. > > No interference, no other programs barging in, and no failures that > can't be explained as insufficient power. I don't know why this run > was different from the first wireshark trace. Did you remove some > program or change a udev rule? > > At any rate, the log and the trace don't point to any hardware or > software errors. The only problem is lack of sunlight. > > Alan Stern > -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
