Re: Sony Clie with USB sync

[Daniel Brodie <daniel brodienet com>]

Hello Lars,

Well, I have tried ttyUSB1 before and it didn't work. I also got the
device originally from /var/log/messages

I do have a simmilar problem with pilot-xfer etc... so this problem
appears to not be limited to gnome-pilot.

I have also tried specifying USB instead of serial but it dosn't work.
(gives me a gtk error about not being able to open display)

before I get it to work with the palm I have to do modprobe visor. I
have no idea how to enable it in modules.conf though (need to read
through the man page ;) )

This is the relevant lines from lsmod:
usbcore           78784   1  [visor usbserial usb-storage hid usb-uhci
ehci-hcd]
usb-uhci               26348   0  (unused)
visor                  11144   0
usbserial              22108   0  [visor]
usb-storage            69332   0

and a few more. But as you can see, they are all up and running before
even attempting to get gpilotd to work.

So, now I have to have the palm in the middle of a sync before starting
gpilotd and I also have to kill gpilotd at th eend before the next sync.
This seems like a problem with the usb device somehow. Off to do some
more reaserch :)

Thank you for all the help.
-Daniel

On Thu, 2003-08-14 at 02:10, Lars Weissflog wrote:
> Hi Daniel,
> 
> well I had similar problems in earlier versions. What makes me wonder is
> that you use /dev/ttyUSB0 (Zero). AFAIK it should be /dev/ttyUSB1 (One),
> at least that is what it is for me. But I remember to have read
> somewhere that _some_ palms/clies/whatever use Zero. 
> 
> 2 the list: Please, could one of the developers or "professionals" jump
> in here?
> 
> As far as I remember the problem you face is that the /dev/ttyUSBx
> entries are created when the corresponding port is activated and that
> happens at the moment you initiate the sync from the clie or its cradle.
> I'm not sure what it takes to make them exist permanent, I had that
> problem, too. But it's fixed. But I still have it with pilot-xfer. There
> I have to open a Kernel-log and watch it carefully, and as soon as the
> visor-usb-module reports the connect, I hit ENTER in another terminal
> where pilot-xfer -p /dev/ttyUSB1 -l is waiting. This usually works. 
> 
> Ah, this brings me to a possible solution. Maybe your visor-usb module
> is compiled as module and gets loaded when needed, and at that time the
> /dev/ttyUSB0 and 1 ports get created. Try preloading the module by
> modules.conf at startup so that it is present all the time. Maybe this
> is what I did. Sorry, can't remember exactly. 
> But a good point to continue is opening an xterm and running
> 
> tail -f /var/log/messages
> 
> this gives for my clie just after hitting hotsync on the device:
> 
> ---8<----
> Aug 14 08:03:31 carryme syslogd 1.4.1#10: restart.
> Aug 14 08:05:11 carryme kernel: hub.c: new USB device 00:07.2-2,
> assigned address 3
> Aug 14 08:05:11 carryme kernel: Manufacturer: Palm, Inc.
> Aug 14 08:05:11 carryme kernel: Product: Palm Handheld
> Aug 14 08:05:11 carryme kernel: usbserial.c: Handspring Visor / Palm 4.0
> / Clié 4.x converter detected
> Aug 14 08:05:11 carryme kernel: visor.c: Handspring Visor / Palm 4.0 /
> Clié 4.x: Number of ports: 2
> Aug 14 08:05:11 carryme kernel: visor.c: Handspring Visor / Palm 4.0 /
> Clié 4.x: port 1, is for Generic use and is bound to ttyUSB0
> Aug 14 08:05:11 carryme kernel: visor.c: Handspring Visor / Palm 4.0 /
> Clié 4.x: port 2, is for HotSync use and is bound to ttyUSB1
> Aug 14 08:05:11 carryme kernel: usbserial.c: Handspring Visor / Palm 4.0
> / Clié 4.x converter now attached to ttyUSB0 (or usb/tts/0 for devfs)
> Aug 14 08:05:11 carryme kernel: usbserial.c: Handspring Visor / Palm 4.0
> / Clié 4.x converter now attached to ttyUSB1 (or usb/tts/1 for devfs)
> Aug 14 08:05:18 carryme kernel: usb.c: USB disconnect on device
> 00:07.2-2 address 3
> Aug 14 08:05:18 carryme kernel: usbserial.c: Handspring Visor / Palm 4.0
> / Clié 4.x converter now disconnected from ttyUSB0
> Aug 14 08:05:18 carryme kernel: usbserial.c: Handspring Visor / Palm 4.0
> / Clié 4.x converter now disconnected from ttyUSB1
> 
> ---->8---
> 
> In addition, if I run gpilotd after killing it once, I get:
> 
> ---8<----
> gpilotd-Message: gnome-pilot 2.0.10 starting...
> gpilotd-Message: compiled for pilot-link version ..
> gpilotd-Message: compiled with [VFS] [USB] [IrDA] [Network] 
> gpilotd-Message: Activating CORBA server
> gpilotd-Message: bonobo_activation_active_server_register = 0
> gpilotd-Message: Watching Cradle (/dev/ttyUSB1)
> gpilotd-Message: setting PILOTRATE=9600
> gpilotd-Message: Cradle Cradle has 0 events
> gpilotd-Message: Instantiating 4 conduits...
> ecalconduit-Message: in calendar's conduit_get_gpilot_conduit
> 
> eaddrconduit-Message: in address's conduit_get_gpilot_conduit
> 
> etodoconduit-Message: in todo's conduit_get_gpilot_conduit
> 
> gpilotd-Message: Instantiated 1 backup conduits, 0 file conduits, 3
> other conduits
> gpilotd-Message: HotSync button pressed, synchronizing pilot
> gpilotd-Message: Pilot ID is 9560, name is Clie, owner is Lars W
> gpilotd-Message: Pilot has 0 entries in restore queue
> gpilotd-Message: Pilot has 0 entries in conduit queue
> {ripped here, a lot of sync info follows until the memo-conduit
> crashes.... another story}
> --->8---
> 
> and here is my lsmod
> Module                  Size  Used by    Not tainted
> ide-cd                 27008   0 (autoclean)
> cdrom                  28640   0 (autoclean) [ide-cd]
> maestro3               24712   2 (autoclean)
> ac97_codec             10208   0 (autoclean) [maestro3]
> soundcore               3460   2 (autoclean) [maestro3]
> irtty                   5568   2 (autoclean)
> irda                   81388   1 (autoclean) [irtty]
> apm                     9504   2 (autoclean)
> nls_iso8859-1           2880   1 (autoclean)
> nls_cp437               4384   1 (autoclean)
> vfat                    9052   1 (autoclean)
> fat                    30200   0 (autoclean) [vfat]
> visor                   8608   1
> usbserial              16512   0 [visor]
> hid                    18624   0 (unused)
> usbmouse                1824   0 (unused)
> mousedev                3808   1
> tulip                  39168   1
> input                   3296   0 [hid usbmouse mousedev]
> uhci                   24552   0 (unused)
> usbcore                61728   1 [visor usbserial hid usbmouse uhci]
> rtc                     5884   0 (autoclean)
> parport_pc             14052   1 (autoclean)
> lp                      6560   0 (autoclean)
> parport                25184   1 (autoclean) [parport_pc lp]
> 
> 
> you should preload uhci (or ohci, depending on your hardware), usbcore,
> usbserial and visor. The rest is machine/application specific stuff and
> may differ for you.
> 
> Let us hear more.....
> Good luck
> 
> Lars
> 
> 
> 
> 
> 
> 
> On Thu, 2003-08-14 at 04:56, Daniel Brodie wrote:
> > Thank you for your help. This is some of the info I gathered while
> > playing around with it a bit more.
> > 
> > I removed the applet. The process didn't die. I tried sending -9 but it
> > was properly hung on IO or parent or somthing. (I have been able to make
> > programs that are unkillable like that, IIRC they were some kind of
> > orphans, either way very weird stuff)
> > 
> > Anyway, the only way it works for me is if I start a sync before I run
> > gpilotd otherwise gpilotddosn't work. Also I have to restart gpilotd
> > between sync's. Currently my pilot uses /dev/ttyUSB0 and I have a
> > symbolic link from /dev/pilot to that. I have to use the serial method
> > in gnome-pilot. If I use the usb method my machine freezes (well the
> > sysreq keys work ;) ) and the scool lock and caps lock start flashing.
> > Notice I DO sync using a usb device, but in gnome-pilot I have to set
> > the type of sync to serial.
> > 
> > Now for some output.
> > 
> > gpilotd not during a sync:
> > gpilotd-Message: gnome-pilot 2.0.10 starting...
> > gpilotd-Message: compiled for pilot-link version 0.11.7
> > gpilotd-Message: compiled with [VFS] [USB] [IrDA] [Network]
> >  
> > (gpilotd:4701): gpilotd-WARNING **: Could not open device Cradle
> > (/dev/ttyUSB0): reason: "No such device".
> > gpilotd-Message: Activating CORBA server
> > gpilotd-Message: bonobo_activation_active_server_register = 0
> > 
> > I try sync'ing but nothing happens, and in the end my palm gives me an
> > error about not connecting.
> > 
> > gpilotd started during a sync:
> > gpilotd-Message: gnome-pilot 2.0.10 starting...
> > gpilotd-Message: compiled for pilot-link version 0.11.7
> > gpilotd-Message: compiled with [VFS] [USB] [IrDA] [Network]
> > gpilotd-Message: Activating CORBA server
> > gpilotd-Message: bonobo_activation_active_server_register = 0
> > gpilotd-Message: Watching Cradle (/dev/ttyUSB0)
> > gpilotd-Message: Woke on Cradle
> > gpilotd-Message: setting PILOTRATE=57600
> > gpilotd-Message: Cradle Cradle has 0 events
> > gpilotd-Message: Instantiating 1 conduits...
> > gpilotd-Message: Instantiated 1 backup conduits, 0 file conduits, 0
> > other conduits
> > gpilotd-Message: HotSync button pressed, synchronizing pilot
> > gpilotd-Message: Pilot ID is 29329, name is Sony Clie, owner is Kevin
> > Burke
> > gpilotd-Message: Pilot has 0 entries in restore queue
> > gpilotd-Message: Pilot has 0 entries in conduit queue
> > backupconduit-Message: Making backup of finger
> > backupconduit-Message: Wrote 1 of 1 records, which is good
> > backupconduit-Message: redbook-07.pdf not modified since last sync
> > {much more here, I cut out}
> > backupconduit-Message: Wordnet medium not modified since last sync
> > gpilotd-Message: Synchronization ended
> > 
> > At this point, if I try to sync again it dosn't work, same as before.
> > Restarting gpilotd repeats itself.
> > 
> > Thank you for your help.
> > -Daniel
> > 
> > On Wed, 2003-08-13 at 02:04, Lars Weissflog wrote:
> > > Hi Daniel,
> > > 
> > > I had severe problems with my Clie and gpilotd, too. They are mostly
> > > gone with the newer releases as those you use. For the
> > > memo-files-conduit, please refer to the archives of this list. There
> > > have been dozens of mails lately dealing with this. To keep it short: It
> > > doesn't work at the moment wit gnome-pilot 2.0.10. The memo stuff is
> > > simply broken. Disable that conduit.
> > > 
> > > For the general connection issue try this: 
> > > 1) Do NOT use the applet, remove it from the tray.
> > > 2) open a shell and kill gpilotd. (You must be able to kill it.
> > > Otherwise you don't use linux but some kind of windoze. Try killing it
> > > by its process-id, i.e. "kill -9 1234" if 1234 is the id, you get that
> > > number from "ps aux"
> > > 3) after killing gpilotd start it by simply typing "gpilotd"
> > > 4) send us the output. Maybe your connection settings are completely
> > > wrong.
> > > 5) try to connect and watch the shell. Evo doesn't have to be running
> > > while you do all this, but it should work if evo runs. 
> > > 
> > > For the todo-crashes I have no idea, sorry.
> > > 
> > > Tell us more.
> > > 
> > > HtH
> > > 
> > > Lars 
> > > 
> > > 
> > > On Wed, 2003-08-13 at 07:18, Daniel Brodie wrote:
> > > > Hello,
> > > > 
> > > > I have been trying to get gnome-pilot to work with my Sony Clie with a
> > > > usb sync. pilot-* and memos all work great (after a little tweaking ;)
> > > > ).
> > > > 
> > > > gnome-pilot version: 2.0.10 from NyQuist repository on Redhat 9
> > > > pilot-link version: 0.11.7 from redhat rawhide.
> > > > 
> > > > setting up the gpilot initially with Serial (didn't work with USB? Is
> > > > this because its 'emulated' with serial in the visor module?) works. But
> > > > after trying to set some conduits (Specifically, the test conduit, the
> > > > evolution todo conduit, and the memo_files conduit, now the gpilotd
> > > > froze and evolution crashs every time I try going into the todo folder.
> > > > 
> > > > Basicly my question is, are there conduits that I can try out that are
> > > > known to work so I can test this to make sure it really works? I
> > > > currently can't even kill -9 gpilotd! (I cant reboot curently but I will
> > > > try to play with things again next time I can)
> > > > 
> > > > Also, there was one time where the sync'ing didnt freeze (I believe only
> > > > the test conduit was enabled) but instead nothing happened, except my
> > > > palm complained that it couldnt connect. (the color on the applet didnt
> > > > change either).
> > > > 
> > > > I would appiciate any help.
> > > > Thank you.
> > > > 	-Daniel
> > > > 
> > > > _______________________________________________
> > > > gnome-pilot-list mailing list
> > > > gnome-pilot-list gnome org
> > > > http://mail.gnome.org/mailman/listinfo/gnome-pilot-list
> > 
> > _______________________________________________
> > gnome-pilot-list mailing list
> > gnome-pilot-list gnome org
> > http://mail.gnome.org/mailman/listinfo/gnome-pilot-list