It was necessary to list all the USB devices (significant research) to find out that the palm device was actually connected to: /dev/ttyUSB1 there is no reason to believe that any remotely mortal user would consider doing this. 

According to the palm device you can not sync with Gnome-Pilot if you have password enabled in the device! This works on other distros! Please fix!

OK OK... My fault on the last one. Please disregard Comment #2. The defaults need to visited though. I cant imagine that a user would want to poke around in the command line and using odd tools to find out what device the system thinks that the pilot is plugged in to. 

Added HAL support for code 10, needs retesting.

Ok, found an issue in pilot-link which I fixed.  Found an issue in gnome-pilot for non-visor USB devices which I have a patch for.

As for searching around for the right device, indeed this is very annoying but until the advent of HAL and udev, there wasn't much that could be done.

That being said, Kay, Greg, I understand that some other distros are shipping a udev rule something like:
BUS="usb" SYSFS{product}="Palm Handheld*" KERNEL="ttyUSB*" MODE="666"
SYMLINK="pilot"

Does this seem like a viable thing to do?

We ship a rule for this:
  KERNEL=="ttyUSB*", SYSFS{product}=="[Pp]alm*Handheld*", SYMLINK+="pilot"

Indeed, I grepped incorrectly.

For me on a Tungsten T /dev/pilot is mapped incorrectly however.  Its mapped to ttyUSB1 instead of ttyUSB0.  A matrix of these mappings is linked from:
http://pilot-link.org/DeviceMatrix

My device is specifically covered in:
http://www.pilot-link.org/node/111

Greg, can anything be done about this?

If so, is there a way then to distinguish the two devices that show up in HAL when the hotsync button is pressed, ie a way to mark with a property what is actually the hotsync device (maybe merge properties based on the device matrix above Kay?).

I have a patch now that resolves symbolic links (for /dev/pilot) via readlink to try and match on properties in hal, however the /dev/pilot link is not absolute, so thats a bit annoying.

*** Bug 103053 has been marked as a duplicate of this bug. ***

No, there is nothing "automatic" that can be done to create this pilot symlink.

We can try to match up some devices, like you are trying to do now, and that's
about the best we can do.  The problem is the device tries to export
attributes about which port is which (debug, hotsync, etc.), but for a 
large number of devices, it lies about this.  This is especially true for 
the Sony Clie devices, which are known to do this (so much so, we have a
special kernel log message telling the user that the device lied about
the ports, so they should try the other one instead.)

Sorry about this, go blame Palm if you want to, it's their fault :)

Ok, submitted a package that works with hal/non-visor devices and older visor devices.  Left /dev/pilot in for those devices it will work for.  A couple as per the matrix above will not work and will need /dev/ttyUSB0.

Really we need to move completely to libusb or better hal info, but is a separate big feature already in the feature database.

Btw, to the reporter, if you have a Tungsten T like me, you'll need to use /dev/ttyUSB0.

This is still broken even when using gnome-pilot-2.0.13-45.  I'm using SUSE 10.1 with all patches and a Treo 650.  I get this output when trying to sync:
>/opt/gnome/lib/gnome-pilot/gpilotd
gpilotd-Message: gnome-pilot 2.0.13 starting...
gpilotd-Message: compiled for pilot-link version 0.11.8
gpilotd-Message: compiled with [VFS] [USB] [IrDA] [Network] 
gpilotd-Message: Activating CORBA server
gpilotd-Message: bonobo_activation_active_server_register = 0
gpilotd-Message: Watching Treo650 (/dev/ttyUSB1)
gpilotd-Message: palm pda detected by HAL
gpilotd-Message: setting PILOTRATE=115200

(gpilotd:589): gpilotd-WARNING **: pi_accept_to: Connection timed out

(gpilotd:589): gpilotd-WARNING **: pi_accept_to: timeout was 2 secs
gpilotd-Message: palm pda detected by HAL
gpilotd-Message: setting PILOTRATE=115200

(gpilotd:589): gpilotd-WARNING **: pi_accept_to: Connection timed out

(gpilotd:589): gpilotd-WARNING **: pi_accept_to: timeout was 2 secs

What does your kernel log messages show when you try to do this?

Nov  1 12:36:14 mn65-eggplant kernel: usb 5-1: new full speed USB device using uhci_hcd and address 4
Nov  1 12:36:14 mn65-eggplant kernel: usb 5-1: new device found, idVendor=0830, idProduct=0061
Nov  1 12:36:14 mn65-eggplant kernel: usb 5-1: new device strings: Mfr=1, Product=2, SerialNumber=5
Nov  1 12:36:14 mn65-eggplant kernel: usb 5-1: Product: Palm Handheld
Nov  1 12:36:14 mn65-eggplant kernel: usb 5-1: Manufacturer: PalmOne, Inc.
Nov  1 12:36:14 mn65-eggplant kernel: usb 5-1: SerialNumber: PalmSN12345678
Nov  1 12:36:14 mn65-eggplant kernel: usb 5-1: configuration #1 chosen from 1 choice
Nov  1 12:36:14 mn65-eggplant kernel: visor 5-1:1.0: Handspring Visor / Palm OS converter detected
Nov  1 12:36:14 mn65-eggplant kernel: usb 5-1: Handspring Visor / Palm OS converter now attached to ttyUSB0
Nov  1 12:36:14 mn65-eggplant kernel: usb 5-1: Handspring Visor / Palm OS converter now attached to ttyUSB1
Nov  1 12:36:55 mn65-eggplant kernel: usb 5-1: USB disconnect, address 4
Nov  1 12:36:55 mn65-eggplant kernel: visor 5-1:1.0: device disconnected
Nov  1 12:36:55 mn65-eggplant kernel: visor ttyUSB0: Handspring Visor / Palm OS converter now disconnected from ttyUSB0
Nov  1 12:36:55 mn65-eggplant kernel: visor ttyUSB1: Handspring Visor / Palm OS converter now disconnected from ttyUSB1

Thanks, nothing looks improper in the kernel log, so it must be an issue higher up the stack.

On comment 13: the patch was never shipped on 10.1, its in SLE 10 products and 10.2.

You can get an updated package at:
http://ftp.osuosl.org/pub/suse/suse/i386/supplementary/GNOME/update_for_10.1/applications/gnome-pilot-2.0.13-50.i586.rpm

Do you have a 64bit version of the package as well?

http://ftp.osuosl.org/pub/suse/suse/x86_64/supplementary/GNOME/update_for_10.1/applications/gnome-pilot-2.0.13-50.x86_64.rpm

Thanks, this is a little better.  I can get it to connect now.  However now I get this error when trying to sync or pull the sync id off the pilot:

(gpilotd:18779): gpilotd-WARNING **: Number of pilots is configured to 0
gpilotd-Message: Watching Treo650 (/dev/ttyUSB1)
gpilotd-Message: corba: get_user_info(cradle=Treo650,survival=0,timeout=0)
gpilotd-Message: assigned handle num 2
gpilotd-Message: corba: get_system_info(cradle=Treo650,survival=0,timeout=0)
gpilotd-Message: assigned handle num 3
gpilotd-Message: palm pda detected by HAL
gpilotd-Message: setting PILOTRATE=115200

(gpilotd:18779): gpilotd-WARNING **: An error occured while getting the pilot's system data
gpilotd-Message: palm pda detected by HAL
gpilotd-Message: setting PILOTRATE=115200

(gpilotd:18779): gpilotd-WARNING **: pi_accept_to: Connection timed out

(gpilotd:18779): gpilotd-WARNING **: pi_accept_to: timeout was 2 secs

I suspect now that not having pilot-link 0.12 is the issue (which fixed some issues for newer pilots).

And where might I find that?  It doesn't appear to be at your previous URL.

I found a tarball of pilot-link and installed that in a temporary directory, then installed a patched version of gnome-pilot in a temporary directory and that combination allowed me to communicate successfully with my palm using gpilotd.  Any chance these versions will be in 10.1?  Any chance someone already has them packaged up in 64bit RPMs?

See the bug here: http://bugzilla.gnome.org/show_bug.cgi?id=321947
The pilot-link package is from pilot-link.org
The gnome-pilot package is from http://www.inference.phy.cam.ac.uk/mcdavey/downloads/gnome-pilot-2.0.14p1.tar.gz

We have recent versions of pilot-link in FACTORY, but we can't mix the new version of pilot-link with older version rpms in 10.1, becasue the API is incompatible with the older version of application which use pilot-link.

I tried getting all of the dependencies, but the list got too long.  So I guess I have to wait until 10.2 is out.  Hopefully it'll work right then, as my Palm hasn't worked since 9.3.