Trying out a brand new Oceanic Geo 2.0 -- can't sync

Benjamin Kahn xkahn at
Fri Jun 28 13:55:57 PDT 2013

Thanks for your help Jef. 

On Fri, 2013-06-28 at 22:27 +0200, Jef Driesen wrote:
> On 28-06-13 21:27, Benjamin Kahn wrote:
> > I just received a brand new Oceanic Geo 2.0 for a dive trip next month.
> > I won't have much network access, so I'd like to make sure I can connect
> > my new dive computer to my Fedora 18 laptop before going.
> In some cases downloading a dive computer without any dives can fail, because 
> this is a rather rare condition, and not all code has ever been tested in such 
> condition. Data from a brand new device is hard to get, because it usually 
> doesn't last long in that condition. Nobody buys a dive computer to stay dry  :-)

I suspected that this might be an issue. It's an inconvenience though
since I can't test the interface before I need it.

> > I've seen two types of errors. Sometimes I immediately get a permission
> > error:
> >
> >          [0.000] ERROR: Device or resource busy (16) [in serial_posix.c:107 (serial_open)]
> >          [0.000] ERROR: Failed to open the serial port. [in oceanic_atom2.c:377 (oceanic_atom2_device_open)]
> Looks like some other applications already has opened the serial port in 
> exclusive mode. That might be modemmanager, as Linus explained in his reply. 
> Trying to shutdown that process is probably a good idea.

Yep. Following Linus's advice, I uninstalled and killed modem-manager.
That fixed the device busy errors. Thank you!

> > And sometimes, I get further. When this happens the DC appears to crash.
> > Every element on the LCD turns black until I press one of the keys on
> > the DC.  When this happens, I see this output to the console:
> >
> >          [0.112] ERROR: Unexpected answer start byte(s). [in oceanic_atom2.c:287 (oceanic_atom2_send)]
> >          [3.226] ERROR: Failed to receive the answer. [in oceanic_atom2.c:281 (oceanic_atom2_send)]
> >          [3.339] ERROR: Unexpected answer start byte(s). [in oceanic_atom2.c:287 (oceanic_atom2_send)]
> >          [3.449] WARNING: Uninitialized logbook entries detected! [in oceanic_common.c:404 (oceanic_common_device_foreach)]
> >          [6.453] ERROR: Failed to receive the answer. [in oceanic_atom2.c:281 (oceanic_atom2_send)]
> The warning you can just ignore. For the errors, I'll need a full debug log. The 
> easiest way to get that is to run the libdivecomputer "universal" test app. You 
> can build it from source, or download a pre-compiled binary here:
> Then run it with these options:
> ./universal -l output.log -d output.xml -n "Oceanic Geo 2.0" -v /dev/ttyUSB0
> and send the output.log file.


> >          Jun 28 00:42:24 localhost mtp-probe: checking bus 1, device 27: "/sys/devices/pci0000:00/0000:00:1a.0/usb1/1-1/1-1.2"
> >          Jun 28 00:42:24 localhost mtp-probe: bus: 1, device: 27 was not an MTP device

> If mtp-probe is probing the device, as it's name indicates, then that might be 
> another candidate interfering with the download. If the probing involves sending 
> data to the device, than that may also explain your dive computer acting weird 
> because it's receiving data it doesn't understand.

It looks like mtp-probe runs when I insert the USB cable, not when I
attach the dive computer to the cable. I believe this means it isn't a
-------------- next part --------------
A non-text attachment was scrubbed...
Name: output.log
Type: text/x-log
Size: 2433 bytes
Desc: not available
URL: <>

More information about the subsurface mailing list