On Thu, 23 Jan 2014 15:40:28 -0500 glphvgacs <darwinskernel@xxxxxxxxx> wrote: > On Thu, Jan 23, 2014 at 03:12:10PM -0500, Alan Stern wrote: > > On Thu, 23 Jan 2014, glphvgacs wrote: > > > > > > The trace shows that the mouse appears, and then apparently is > > > > unused. It kind of looks as though the mouse goes into runtime > > > > suspend, but not entirely. After 60 seconds, the mouse > > > > disconnects itself and then reconnects a little later. > > > > > > > > Does the mouse work at all? From what I can tell in the usbmon > > > > trace, nothing would happen if you moved it or pressed a button. > > > > > > works just fine. > > > > That's strange. Can you provide a usbmon trace that shows what > > happens when you use the mouse? And to keep down the size of the > > trace file, can you unplug the USB drive (the sdb drive) while > > running the test? > > nope, i'm booting off of that hd (IDE) that's hooked up to usb and > yes i caught a lot of it's traffice since it was on bus2. > could it have something to do with this? i'm moving to usb3 soon > though. did you have a change to look at the systemd-udev msg in my > previous posts? Dear glphygacs, You might like to read the kernel archives for "Repeated disconnects of wired mouse" where Alan and I tried to work out why my wired mice disconnected with perfect regularity. I tried different mice made by different manufacturers (but with the same chip, it turned out) in different computers running different linux kernel versions and options. I never did find the cause. The way I worked around it was like this: I first noticed the problem on a computer which used USB extenders to send the USB signals to another room. Thinking it was an electrical noise problem I tried out cheap extenders and expensive extenders to no avail. However, one of the peripherals on the USB was a mains-powered scanner. When I put it on a separate bus a lot of the disconnections went away. My particular electrical arrangements might well have involved a large ground loop which would pick up interference. Although I doubt this is your problem, be aware that some equipment can be dc-coupled and therefore invite ground loops. VGA extenders are the worst. I've had one which spewed wide-spectrum interference all over the mains supply and could be picked up with a slightly off-station portable radio from hundreds of yards away. My problem continued when I connected the same mice to a laptop running on battery in the middle of a field, so it wasn't just down to electrical noise. In fact the only way I was able to resolve the problem was to use a wireless mouse. This has proved to be almost 100% reliable in operation since then. I concluded that somehow I'd managed to damage a bunch of wired mice. I was going to throw them out but as it happens a couple of days ago I re-discovered them and thought I'd connect them (via an extender) to one of the computers I'd previously had problems with. I've not had any disconnections at all. I'm now using kernel 3.10.3. I'm baffled by all of this. Perhaps I was suffering from a lot of interference from a 3rd party at the time. Perhaps it's moved over to you! Who can say? Nevertheless, I hope I've given you some food for thought. As a point of interest can you tell me what your motherboard is (ie. what the USB chip is), what is the name of the chip in your mouse? Yours, Larry. -- 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
