[jp] /dev/pilot

  • Previous message: [jp] /dev/pilot
  • Next message: [jp] /dev/pilot
  • David A. Desrosiers desrod at gnu-designs.com
    Mon Jan 29 09:02:23 EST 2007

     

    > I know it works as it sees the pen drive. Funny thing.... it was 
    > seeing my card reader but now it does not.
    
    This might be a clue to a lower-level problem... more in a moment.
    
    >> 2. The proper usb support was not loaded (uhci, ehci, ohci)
    
    > modprobe says I have none of those however I tried it with kpilot 
    > and kpilot auto-detected the settings
    
    First part: Did you do a "modprobe uhci_hcd && modprobe ehci_hcd && 
    modprobe ohci_hcd;"? This is a brute-force method to help with 
    testing, but it will at least provide the lower-level usb support your 
    hardware needs to recognize the Palm device.
    
    Second part: Bam! Did you remember to kill the kpilotDaemon process 
    after you "tried" KPilot, and before you ran J-Pilot? KPilot has a 
    daemon process that "listens" for connections on the standard Palm 
    devices, and if this is running, you'll block access to the same 
    device node that J-Pilot needs to communicate with. The same problem 
    exists with gnome-pilot and the gnome-pilot daemon process (called 
    gpilotd).
    
    Standard GNOME desktops have an option in the control-center called 
    "PalmOS Devices", and many users click on this, thinking it will 
    configure the device connection "globally", so any application that 
    wants to talk to a Palm device, can. This is incorrect.
    
    What this option DOES do, is configure gnome-pilot, and launches 
    gpilotd into the background (daemonizing its process), so now you 
    can't use any other application (KPilot, pilot-link, J-Pilot, etc.) 
    without killing this process off _manually_ first. Its a bit of a 
    misnomer, and needs to have a warning dialog set up when launched:
    
     	"WARNING: This will only configure gnome-pilot.
     	 If you choose to run another Palm GUI application,
     	 please do not configure it via this mechanism."
    
    ...or something along those lines.
    
    >> 3. The visor driver does not know your device type
    
    > it worked before I put the new kernel on
    
    Another clue here. Sounds like you have broken, misconfigured or bogus 
    kernel. Why did you upgrade the kernel? What wasn't working that 
    necessitated this upgrade? Remember, this isn't Windows, upgrades 
    aren't "necessary" unless they fix bugs, security issues or add device 
    functionality that you need.
    
    If your card reader worked before, and no longer does... if your visor 
    module detected your device before, and no longer does, I'd back that 
    kernel down to the version that DID work in those circumstances, or 
    check the Mepis kernel bugtracker and see if anyone has reported these 
    issues, fixed them, or supplied an updated kernel to address them.
    
    
    David A. Desrosiers
    desrod at gnu-designs.com
    http://gnu-designs.com
    
    

     

     

    More information about the jpilot mailing list