opensync-users — The place to get help for the OpenSync project

Re: [Opensync-users] syncing events with SE K550i

[Benjamin G. <be...@gm...>]

Hi,

On Tuesday 29 September 2009 12:44 you wrote:
> Hi,
>
> is there a way to sync events with the Sony Ericsson K550i? All I get
> if this:
>
> $ osynctool --sync tel
> Synchronizing group "tel"
> The previous synchronization was unclean. Slow-syncing
> Main sink of member 3 of type syncml-obex-client just connected

I have not done it in a while, but I had success syncing my K550i with the 
irmc plugin instead of syncml-obex. I believe to remember that somehow you 
have to exclude the notes data type, too (although I could confuse that 
with my Windows Mobile (synce) device).

That was some months ago with opensync-0.22. I currently have no access to 
look at the configuration used back then.

Sorry for the incomplete information, but I hope it helps.

Ben

Re: [Opensync-users] syncing events with SE K550i

[Tino K. <tin...@ti...>]

On Tue, Sep 29, 2009 at 22:34:54 +0200, Christopher Stender wrote:
> Am Tuesday 29 September 2009 21:44:16 schrieb Tino Keitel:
> > Hi,
> >
> > is there a way to sync events with the Sony Ericsson K550i?
> 
> Please try "<Name>Agenda</Name>" instead of "<Name>Calendar</Name>".

This doesn't change anything.

Regards,
Tino

Re: [Opensync-users] syncing events with SE K550i

[Christopher S. <cst...@su...>]

Am Tuesday 29 September 2009 21:44:16 schrieb Tino Keitel:
> Hi,
>
> is there a way to sync events with the Sony Ericsson K550i?

Please try "<Name>Agenda</Name>" instead of "<Name>Calendar</Name>".

Best,
Christopher
-- 
Christopher Stender, R&D Preload Department
SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
Maxfeldstraße 5, 90409 Nürnberg, Germany

[Opensync-users] syncing events with SE K550i

[Tino K. <tin...@ti...>]

Hi,

is there a way to sync events with the Sony Ericsson K550i? All I get
if this:

$ osynctool --sync tel
Synchronizing group "tel" 
The previous synchronization was unclean. Slow-syncing
Main sink of member 3 of type syncml-obex-client just connected
contact sink of member 1 of type evo2-sync just connected
event sink of member 1 of type evo2-sync just connected
Main sink of member 1 of type evo2-sync just connected
event sink of member 3 of type syncml-obex-client just connected
contact sink of member 3 of type syncml-obex-client had an error:
Timeout.
The sync failed: At least one object engine failed while connecting.
Aborting
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/plugin/opensync_context.c:164:E:osync_context_report_slowsync:
/ Assertion "context" failed
Aborted (core dumped)

Just contacts work fine. If I add an event resource to
syncml-obex-client.conf, the above error occurs.

Regards,
Tino

Re: [Opensync-users] osynctool --sync hangs when using bluetooth

[Graham C. <g+o...@co...>]

On Tuesday 29 September 2009 12:01:18 Michael Bell wrote:
> This bug exists for a long time and I think that the limit is not
> configurable today. So the bug must be fixed (e.g. by a configurable
> limit which is much bigger than 5 by default).

I will check in a workround (either disable the check or make it a large 
number) today.  A real fix will take a while longer.

Graham

Re: [Opensync-users] osynctool --sync hangs when using bluetooth

[Michael B. <mic...@cm...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Tino,

Tino Keitel wrote:
> 
> finally, I figured out how to transfer new contacts to the phone. It
> looks like there is some kind of limit. If I add just 3 contacts and
> not over 100, it works. I also tried with 10 and 5 contacts, but
> osynctool still hangs.
> 
> Can I enforce this limit in the config file somehow?

This is an old bug of OpenSync internal change handling.

OpenSync's engine has an internal pendingLimit for the number of changes
which are pushed to a plugin. If this limit is reached then OpenSync
waits until at minimum one change is committed.

libsyncml as SyncML implementation only sends a message if the message
is full in terms of MaxMsgSize. This is usually a limit of the phone.

So if the limit of the phone (MaxMsgSize) would allow to pack 6 changes
in a message and the pendingLimit of OpenSync is 5 then there is a dead
lock. OpenSync waits for a commit signal for at minimum one change and
libsyncml waits for the next change.

This bug exists for a long time and I think that the limit is not
configurable today. So the bug must be fixed (e.g. by a configurable
limit which is much bigger than 5 by default).

Actually I'm not able to fix it because my actual knowledge of the
related OpenSync code is too small :(

Best regards

Michael
- --
___________________________________________________________________

Michael Bell                        Humboldt-Universitaet zu Berlin

Tel.: +49 (0)30-2093 2482           ZE Computer- und Medienservice
Fax:  +49 (0)30-2093 2704           Unter den Linden 6
mic...@cm...       D-10099 Berlin
___________________________________________________________________

PGP Fingerprint: 09E4 3D29 4156 2774 0F2C  C643 D8BD 1918 2030 5AAB
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkrB6PsACgkQ2L0ZGCAwWquVNQCfcqpnMFEAlEopRyT9zegJl5KW
efAAoJ1yXRlyDh4dfMuHtgT4skZLKX+j
=5uRo
-----END PGP SIGNATURE-----

[Opensync-users] Segfault at slow sync

[Tino K. <tin...@ti...>]

Hi,

I get a segmentation fault in opensync 0.39 when running
osynctool --sync tel --conflict 1:

Synchronizing group "tel" 
The previous synchronization was unclean. Slow-syncing
Main sink of member 3 of type syncml-obex-client just connected
contact sink of member 1 of type evo2-sync just connected
Main sink of member 1 of type evo2-sync just connected
contact sink of member 3 of type syncml-obex-client just connected
All clients connected or error
Main sink of member 3 of type syncml-obex-client just sent all changes
Received an entry pas-id-4AC1C32100000025 (xmlformat-contact) from
member 1 (evo2-sync). Changetype ADDED
Received an entry pas-id-4AC1C3210000002E (xmlformat-contact) from
member 1 (evo2-sync). Changetype ADDED
...
Received an entry pas-id-4AC1C32200000063 (xmlformat-contact) from
member 1 (evo2-sync). Changetype ADDED
Received an entry pas-id-4AC1C3220000006C (xmlformat-contact) from
member 1 (evo2-sync). Changetype ADDED
contact sink of member 1 of type evo2-sync just sent all changes
Main sink of member 1 of type evo2-sync just sent all changes
Received an entry 890 (xmlformat-contact) from member 3
(syncml-obex-client). Changetype ADDED
Received an entry 861 (xmlformat-contact) from member 3
(syncml-obex-client). Changetype ADDED
...
Received an entry 918 (xmlformat-contact) from member 3
(syncml-obex-client). Changetype ADDED
Received an entry 899 (xmlformat-contact) from member 3
(syncml-obex-client). Changetype ADDED
contact sink of member 3 of type syncml-obex-client just sent all
changes
All clients sent changes or error
Conflict for Mapping 0x7f68a010fe40: Solving conflict
Conflict for Mapping 0x7f68a0138a50: Solving conflict
...
Conflict for Mapping 0x7f68a0148200: Solving conflict
Conflict for Mapping 0x7f68a0148340: Solving conflict
All changes got mapped
All conflicts have been reported
All changes got multiplied
Segmentation fault (core dumped)

Here is the backtrace, I hope it is usefull:

(gdb) bt
#0  xmlDocDumpFormatMemoryEnc__internal_alias
# (out_doc=0x540a0d3b3b3b616f, doc_txt_ptr=0x7f68ab589c48, 
    doc_txt_len=0x7f68ab589c54, txt_encoding=0x0, format=1) at
xmlsave.c:2202
#1  0x00007f68b5b38784 in osync_xmlformat_assemble (xmlformat=<value
# optimized out>, buffer=0x7f68ab589c48, 
    size=0x7f68ab589c54, error=<value optimized out>)
    at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/xmlformat/opensync_xmlformat.c:251
#2  0x00007f68b330c08a in conv_xmlformat_to_vcard (
    input=0x7f68a0019e20
"BEGIN:VCARD\r\nVERSION:2.1\r\nN:Restaurant;Goa;;;\r\nTEL;WORK:08921111789\r\nEND:VCARD\r\n", 
    inpsize=<value optimized out>, output=0x7f68ab589ca8,
outpsize=0x7f68ab589cc0, free_input=0x7f68ab589cbc, 
    config=0x7f68a000c970 "", error=0x7f68ab589e80, target=0)
    at
/home/scratch2/src/opensync/0.39/libopensync-plugin-vformat-0.39/src/xmlformat-vcard.c:1909
#3  0x00007f68b5b14f23 in osync_converter_invoke (converter=0x7d6dc0,
/data=0x7f68a0062250, config=0x7f68a000c970 "", 
    error=0x7f68ab589e80) at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/format/opensync_converter.c:193
#4  0x00007f68b5b1660c in osync_format_env_convert (env=<value
/optimized out>, path=0x7f68a0071f00, data=0x7f68a0062250, 
    error=0x7f68ab589e80) at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/format/opensync_format_env.c:1213
#5  0x00007f68b5b0f93e in osync_entry_engine_convert
/(entry_engine=0x7f68a013ea00, formatenv=0x7d3ed0,
/objtype_sink=0x62ee60, 
    cachedpath=0x7f68ab589db0, error=0x7f68ab589e80)
    at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_mapping_entry_engine.c:252
#6  0x00007f68b5b1363c in osync_sink_engine_convert_to_dest
/(engine=<value optimized out>, formatenv=0x7d3ed0, 
    error=0x7f68ab589e80) at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_sink_engine.c:196
#7  0x00007f68b5b10430 in osync_obj_engine_prepare_write
/(engine=0xcc1af0, error=0x7f68ab589e80)
    at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_obj_engine.c:1459
#8  0x00007f68b5b12c83 in osync_obj_engine_command (engine=0xcc1af0,
/cmd=<value optimized out>, error=0x7f68ab589e80)
    at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_obj_engine.c:1208
#9  0x00007f68b5b0bf1d in osync_engine_event (engine=0x7b3020,
/event=<value optimized out>)
    at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_engine.c:1989
#10 0x00007f68b5b0df65 in _osync_engine_generate_multiplied_event
/(objengine=<value optimized out>, 
    event=<value optimized out>, error=0x0, userdata=<value optimized
out>)
    at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_engine.c:1153
#11 _osync_engine_generate_event (objengine=<value optimized out>,
/event=<value optimized out>, error=0x0, 
    userdata=<value optimized out>) at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_engine.c:1481
#12 _osync_engine_event_callback (objengine=<value optimized out>,
/event=<value optimized out>, error=0x0, 
    userdata=<value optimized out>) at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_engine.c:1515
#13 0x00007f68b5b10811 in osync_obj_engine_event (engine=0xcc1af0,
/event=OSYNC_ENGINE_EVENT_MULTIPLIED, error=0x0)
    at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_obj_engine.c:1306
#14 0x00007f68b5b12d42 in osync_obj_engine_command (engine=0xcc1af0,
/cmd=<value optimized out>, error=0x7f68ab589f68)
    at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_obj_engine.c:1203
#15 0x00007f68b5b0d03d in osync_engine_command (engine=0x7b3020,
/command=0x7f68a02296a0)
    at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_engine.c:1828
#16 0x00007f68b5b0d3eb in _command_dispatch (source=<value optimized
/out>, callback=<value optimized out>, 
    user_data=<value optimized out>) at
/home/scratch2/src/opensync/0.39/libopensync-0.39/opensync/engine/opensync_engine.c:365
#17 0x00007f68b585612a in g_main_context_dispatch () from
//lib/libglib-2.0.so.0
#18 0x00007f68b5859988 in ?? () from /lib/libglib-2.0.so.0
#19 0x00007f68b5859e5d in g_main_loop_run () from /lib/libglib-2.0.so.0
#20 0x00007f68b587ef84 in ?? () from /lib/libglib-2.0.so.0
#21 0x00007f68b4044f9a in start_thread () from /lib/libpthread.so.0
#22 0x00007f68b559656d in clone () from /lib/libc.so.6
#23 0x0000000000000000 in ?? ()

Regards,
Tino

Re: [Opensync-users] osynctool --sync hangs when using bluetooth

[Tino K. <tin...@ti...>]

Hi,

the limit seems to be somewhere between 2054 (hangs) and 2004 bytes
(works). Maybe 2048?

Regards,
Tino

Re: [Opensync-users] osynctool --sync hangs when using bluetooth

[Tino K. <tin...@ti...>]

Hi,

finally, I figured out how to transfer new contacts to the phone. It
looks like there is some kind of limit. If I add just 3 contacts and
not over 100, it works. I also tried with 10 and 5 contacts, but
osynctool still hangs.

Can I enforce this limit in the config file somehow?

Regards,
Tino

Re: [Opensync-users] osynctool --sync hangs when using bluetooth

[Tino K. <tin...@ti...>]

Hi,

here is some further information from sdptool:

$ sdptool search --bdaddr 00:1C:A4:BA:26:68 SYNCML
Searching for SYNCML on 00:1C:A4:BA:26:68 ...
Service Name: OBEX SyncML Client
Service RecHandle: 0x10001
Service Class ID List:
  UUID 128: 00000002-0000-1000-8000-0002ee000002
Protocol Descriptor List:
  "L2CAP" (0x0100)
  "RFCOMM" (0x0003)
    Channel: 1
  "OBEX" (0x0008)

My opensync config is attached.

Any help would be appreciated.

Regards,
Tino

Re: [Opensync-users] osynctool --sync hangs when using bluetooth

[Tino K. <tin...@ti...>]

> On Mon, Sep 28, 2009 at 22:48:47 +0200, Tino Keitel wrote:

[...]

> Hi,
> 
> I forgot to mention that I use a SE k550i mobile phone. I just saw that
> reading contacts from the phone works. But as soon as opensync tries to
> write something to the phone (saving new contacts, deleting existing
> contacts) osynctool hangs.

I can also modify a contact on the phone, run osynctool --sync, and the
modified contact will be transfered into evolution.

I can also copy the vcf file with all the contacts to the phone using
gnome-obex-send. Only opensync fails.

Regards,
Tino

Re: [Opensync-users] osynctool --sync hangs when using bluetooth

[Tino K. <tin...@ti...>]

On Mon, Sep 28, 2009 at 22:48:47 +0200, Tino Keitel wrote:
> Hi,
> 
> I set up a sync group with evo2-sync and file-sync members, and syncing
> worked.
> 
> Then I modified the setup to sync evo2-plugin and syncml-obex-client
> over bluetooth. Then I get this output:
> 
> --------------------------------------------------------------------
> ...
> Received an entry pas-id-4AC0E6910000005B (xmlformat-contact) from
> member 1 (evo2-sync). Changetype ADDED
> contact sink of member 1 of type evo2-sync just sent all changes
> Main sink of member 1 of type evo2-sync just sent all changes
> contact sink of member 3 of type syncml-obex-client just sent all
> changes
> All clients sent changes or error
> All changes got mapped
> All conflicts have been reported
> All changes got multiplied
> All changes got prepared for write
> 
> Synchronization Forecast Summary:
> 
> ObjType: contact
>         Member 1: Adding(0) Modifying(0) Deleting(0)
>         Member 3: Adding(103) Modifying(0) Deleting(0)
> 
> Do you want to continue the synchronization? (N/y): y
> 
> OK! Completing synchronization!
> contact sink of member 1 of type evo2-sync committed all changes.
> Main sink of member 1 of type evo2-sync committed all changes.
> -------------------------------------------------------------------
> 
> If I type "y", osynctool hangs, and no bluetooth traffic can be seen in
> hcidump. When I try to open the contact list, it gives an error about

Hi,

I forgot to mention that I use a SE k550i mobile phone. I just saw that
reading contacts from the phone works. But as soon as opensync tries to
write something to the phone (saving new contacts, deleting existing
contacts) osynctool hangs.

Regards,
Tino

[Opensync-users] osynctool --sync hangs when using bluetooth

[Tino K. <tin...@ti...>]

Hi,

I set up a sync group with evo2-sync and file-sync members, and syncing
worked.

Then I modified the setup to sync evo2-plugin and syncml-obex-client
over bluetooth. Then I get this output:

--------------------------------------------------------------------
...
Received an entry pas-id-4AC0E6910000005B (xmlformat-contact) from
member 1 (evo2-sync). Changetype ADDED
contact sink of member 1 of type evo2-sync just sent all changes
Main sink of member 1 of type evo2-sync just sent all changes
contact sink of member 3 of type syncml-obex-client just sent all
changes
All clients sent changes or error
All changes got mapped
All conflicts have been reported
All changes got multiplied
All changes got prepared for write

Synchronization Forecast Summary:

ObjType: contact
        Member 1: Adding(0) Modifying(0) Deleting(0)
        Member 3: Adding(103) Modifying(0) Deleting(0)

Do you want to continue the synchronization? (N/y): y

OK! Completing synchronization!
contact sink of member 1 of type evo2-sync committed all changes.
Main sink of member 1 of type evo2-sync committed all changes.
-------------------------------------------------------------------

If I type "y", osynctool hangs, and no bluetooth traffic can be seen in
hcidump. When I try to open the contact list, it gives an error about
another application which uses the contact list. This error goes away
if I abort osynctool.

Member 1 is evolution, member 3 is the mobile phone. I use opensync
0.39.

Regards,
Tino

[Opensync-users] Can OpenSync sync two hsqldb schemas

[J.V. <jv...@gm...>]

Can OpenSync (or SyncML) be used to sync two hsqldb databases?

We are seeking a free or commercial product / api / tool that can be 
embedded
into a desktop Java app to sync two HSQLDB databases and handle complex 
rules
and would like to know if OpenSync is up to the task.

The hsqldb DB has about a dozen tables in a relational schema.

---
Summary:

We have a desktop application that uses the db on the back end to store
info and have some unique syncing problems.

The user has the ability to install the app, do some stuff that puts
data into the local db (call this Client A) and then sync to a central
server, which merely puts the db file on the server (puts the db file
into a column on production database)

The user then has the ability to go to another computer, install the
app, and if the user has an account and logs in, it will download the db
file from a server (where it was put by Client A in the step above)

The user may run either client in offline mode where data is changed at
any point, such that local changes do not get sync'd to the server (but
at some point in the future may get sync'd)

---
What we want to do is handle all cases where Client A (that has been in
offline mode for some time and has made many local changes) gets data
from a recent sync from Client B and handles everything correctly.

At first look this will require a lot of programming to do this, and we
would like to know if there is a tool or product out there that will do
what we need to do.

Is there a tool that can be used to sync between the two databases
(through using a central always on server) when only one app can be
active at any time and handle the offline scenarios?

---
Simply fetching the database files each time a client launches is not an
option, because of the "offline mode" where Client A may have some data
that was added or changed offline (at a later date than the last sync)
so if we get the db file from the server it will erase those changes.

We need to handle complex cases like

Client A: (online mode)
   launch and add some data
   sync the db file to the server
   close client A

Client B: (online mode)
   launch the app and get latest data file from the server
   add some data
   * rename an item
   sync the new data to the server

Client A: (offline mode)
       launch the app, can not sync from server because offline
       * rename the same item as above (so this is the latest rename)
       Now sync from the server

   Expected:  Do not want to get the item renamed as set from Client B,
but should be the rename from Client A, after the sync.

   For this reason we can not simply download the latest db file from
the server and use it.

   We have to do a merge of the db files but handle changes according
to a time stamp, and only merge changes that are the most recent changes.

   There will be collisions and conflicts that we have to resolve and
this looks like a complex programming problem.

If anyone has any idea on free or commercial products that can be
embedded into a Java app to sync two hsqldb databases and handle complex
rules, please
drop me a line.

We have a few ideas, but want to explore all options.

Re: [Opensync-users] OpenSync release 0.39

[Tino K. <tin...@ti...>]

On Mon, Sep 21, 2009 at 18:38:37 +0200, Bj?oern Ricks wrote:
> Hi everyone,
> 
> after a long time we finally released OpenSync 0.39. This will be the

Great!

Is there any documentation for the vformat plugin? At least I'd like to
know what I can expect from it. I have a large .vcf file with contacts,
and a .ics file with dates, which I'd like to sync with by cellphone.
Can I use these files with the libvformat plugin? If yes, how do I set
it up?

Regards,
Tino

[Opensync-users] OpenSync release 0.39

[Bjšo. R. <bjo...@go...>]

Hi everyone,

after a long time we finally released OpenSync 0.39. This will be the
last experimental release before 0.40. You can find the details in the
release notes.

Release Notes:


Yet another E.X.P.E.R.I.M.E.N.T.A.L. release of OpenSync. This release
is intended for developers, testers and for packages. OpenSync 0.3x
releases aren't intended to be stable. We don't recommend this release
for productive use. We don't recommend this release to be packaged as
part of a distribution.

--- end of sad/scary disclamer ---

This is our last release until the next stable release. The complete API
was examined and should be stable.
Changes since last release:

    * Updates and improvements in the API documentation based on doxygen
(ticket #755). Moved doxygen annotations from the source files to the
headers (ticket #898)
    * Shrinked the public API. Less public interfaces for plugins and
applications.
    * Removed XML format from OpenSync core. Now XML format is
maintained separately
    * Reference counting for all OpenSync modules (#978)
    * Sync plugin API changes
          o Introduce optional plugin sink function "connect_done". This
plugin function gets called after connect() and before get_changes().
          o Removed "write" function. This plugin function was never
used inside the engine and just causes confusion. Currently there is
also no need for a "sync framework" to do "single writes" out of sync order
    * Format plugin API changes
          o Implemented optional "validation" format plugin function.
This function is intended to get called after each format conversion
which end in the format plugin specific format, which provide such
validation function (#910)
          o Introduced osync_objformat_set_finalize_func and
osync_objformat_set_initialize_func for the public API. Those function
allow to register format-plugin specific initalize/finalize functions.
Which get called when the format-plugin get loaded by the
format-environment. This function allow to allocated user-data which
will be passed with all format-plugin calls. And must get released
within the finalize call.
          o Changed all format specific function signatures to pass a
void *user_data pointer. Now it is possible to create format specific
data in initialize which gets passed to all other format functions (#1019)
          o Introduced caps_converter function osync_bool (*
OSyncFormatCapsConverterFunc) (OSyncCapabilities *oldcaps,
OSyncCapabilities **newcaps, OSyncError **error); (#1146)
          o Dropped merge/demerge function. Now merge/demere is
independent of formats (#1084)
    * New IPC implementation by Graham Cobb
    * Support for ObjType mixing in OSyncEngine (#992)
    * Multiply callback for synchronization forecast (#850)
    * Format independent use of merge/demerge via new OSyncCapabilities
and OSyncMerger (#1084)
    * Common list pattern for all OpenSync modules (#975)
    * Dropped file location from OSyncHashtable interface (#1082 and #1083)
    * Dropped file location from OSyncAnchor interface
    * Key=value pattern for OSyncAnchor (#1090)
    * Replace OSyncAnchor with OSyncStateDB (#1091)
    * Improved exception handling in all OpenSync modules (#1087)
    * Started to write new OpenSync API usage examples for 0.40 API (#972)

Next steps:

    * Bugfixing: Open Tickets for 0.40
    * Testing
    * Review the internal API
    * Complete the API documentation
    * Finishing OpenSync Whitepaper
    * Porting other plugins to latest changes for 0.3X

Release tarballs:

 http://opensync.org/download/releases/0.39/

Thanks to everyone for actively contributing to the OpenSync project by
testing, writing code, updating tickets and keeping the project alive!

Re: [Opensync-users] configuration a serial usb modem

[Ulrich B. <ubr...@t-...>]

Am Mittwoch, den 26.08.2009, 15:15 +0400 schrieb Olleg Samoylov:
> Some telephones, Samsung for instance, synchronize by serial usb modem.
> Here is example of success command line:
> 
> syncml-obex-client -s /dev/ttyACM0 --slow-sync text/x-vcard contacts
> --wbxml
> 
> But how this can be configured in xml config? What is the "type" number
> for such connection and the xml tag for "-s" parameter?

After looking into the sources, I would think it is the parameter
"DeviceNode". If you have the sources, see the file
opensync_plugin_config.c. I could not find a list of values for the
parameter type.

Do you have a Samsung phone that syncs with the syncml-obex-client?
Which phone is it? I have a Samsung phone SGH-E250i which does not
support syncml.

Best regards

Ulrich

[Opensync-users] configuration a serial usb modem

[Olleg S. <ol...@mi...>]

Some telephones, Samsung for instance, synchronize by serial usb modem.
Here is example of success command line:

syncml-obex-client -s /dev/ttyACM0 --slow-sync text/x-vcard contacts
--wbxml

But how this can be configured in xml config? What is the "type" number
for such connection and the xml tag for "-s" parameter?
-- 
Olleg Samoylov <ol...@mi...>
Moscow Institute of Physics and Technology

[Opensync-users] libsyncml help

[Douglas <dou...@gm...>]

Hi,

I must implement a SyncML module on a mail service based on E-GroupWare.
This synchronization module must act as a server, synchronizing a user
contact list and calendar  with a client application in a mobile device. All
the proccess must be done using just HTTP protocol. I pretend to use as a
start point an implementation that already exists on the SynML server,
adapting and customizing only the necessary.

I need some information about libsyncml project, like implementation
resources and limitations. Could you help me with that? Here are my doubts:

- If exists some funcionalities in libsyncml that was not implemented, in
relation to the protocol specification;

- Support for detection and control of synchronization conflicts;

- Capability of libsyncml to handle device memory limitation, like MaxMem
attributes in the DevInf standard and large object support;

- Level of abstratction and difficulty, to use with other languages, in this
case, PHP.

Regards,

Douglas

Re: [Opensync-users] libsyncml internal error unknown response

[Ulrich B. <ubr...@t-...>]

Am Mittwoch, den 19.08.2009, 17:55 +0200 schrieb Michael Bell:

> First we need the correct AT command sequence. So do you know AT$TSSPCSW=1?
> 
The AT command was send in the windows software just before the obex
communication started. So I think it is equivalent to AT+CPROT=0.

I found the AT-command also on this side:
http://tools.assembla.com/zsharp/wiki/AT
Samsung Z series AT command set

AT$TSSPCSW=1    Start OBEX connection 
AT$TSSPCSW=0    Terminate OBEX connection


> 
> P.S. I go on holidays at Friday. So I think we cannot fix the issue
> until September.

This is OK.

Ulrich

Re: [Opensync-users] libsyncml internal error unknown response

[Michael B. <mic...@cm...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ulrich Brammer wrote:
> 
> OK, you are right, a short Put sequence was send. but it was quite
> different from the one in windows with the authentication enclosed.
> I looked into the communication of libsyncml and my phone with strace

Now it starts to be interesting ;)

> write(10, "ATZ\r"..., 4)                = 4
> select(11, [10], NULL, NULL, {1, 0})     = 1 (in [10], left {0, 852000})
> read(10, "ATZ\r"..., 100)               = 4
> select(11, [10], NULL, NULL, {1, 0})     = 1 (in [10], left {0, 996000})
> read(10, "\r\nOK\r\n"..., 96)           = 6
> write(10, "AT+CGMI\r"..., 8)            = 8
> select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {0, 996000})
> read(10, "AT+CGMI\r"..., 100)           = 8
> select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {1, 0})
> read(10, "\r\nSAMSUNG\r\n"..., 92)      = 11
> select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {1, 0})
> read(10, "\r\nOK\r\n"..., 81)           = 6
> write(10, "AT+CGMM\r"..., 8)            = 8
> select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {0, 996000})
> read(10, "AT+CGMM\r"..., 100)           = 8
> select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {0, 996000})
> read(10, "\r\nE250i\r\n"..., 92)        = 9
> select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {1, 0})
> read(10, "\r\nOK\r\n"..., 83)           = 6

This stuff is performed to get the necessary informations from the
mobile. We could implement every other AT command to get informations
from the mobile if we need it.

> write(2, "\n** (process:5310): WARNING **: U"..., 96) = 96

This should be the warning about an unknown Samsung mobile.

> write(10, "AT$TSSPCSW=1\r"..., 13)      = 13 
> // I changed the normal AAT+CPROT=0 to this command, as it was used in
> the windows software.
> select(11, [10], NULL, NULL, {1, 0})     = 1 (in [10], left {0, 996000})
> read(10, "AT$TSSPCSW=1\r"..., 100)      = 13
> select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {1, 0})
> read(10, "\r\nCONNECT\r\n"..., 87)      = 11

Any ideas what AT$TSSPCSW=1 means?

> // OK you are right, here is a PUT sequence. The windows software
> contains here the authentication sequence from the server.
> write(10, "\202\0\337\313\0\0\0\2B\0 application/vnd.syncml"..., 223) =
> 223
> select(11, [10], NULL, NULL, {20, 0})   = 1 (in [10], left {20, 0})
> read(10, "\240\0\3"..., 512)            = 3
> write(10, "\203\0(\313\0\0\0\2B\0 application/vnd.syncml"..., 40) = 40
> select(11, [10], NULL, NULL, {20, 0})     = 1 (in [10], left {19,
> 952000})
> // the phone E250i does not accept here. A phone E250 does not answer at
> all at this place.

The mobile does not answer and this means our message is wrong. So the
libsyncml OBEX code generally work we have just to modify it for your
Samsung E250.

First we need the correct AT command sequence. So do you know AT$TSSPCSW=1?

Michael

P.S. I go on holidays at Friday. So I think we cannot fix the issue
until September.
- --
___________________________________________________________________

Michael Bell                        Humboldt-Universitaet zu Berlin

Tel.: +49 (0)30-2093 2482           ZE Computer- und Medienservice
Fax:  +49 (0)30-2093 2704           Unter den Linden 6
mic...@cm...       D-10099 Berlin
___________________________________________________________________

PGP Fingerprint: 09E4 3D29 4156 2774 0F2C  C643 D8BD 1918 2030 5AAB
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkqMIGQACgkQ2L0ZGCAwWquhzgCdHx6PkjiMSvlOHf0uoryw0gFV
M/8AmwXzdhiJQdYRRLl0Qn5Y3iBLdZhO
=RD9j
-----END PGP SIGNATURE-----

Re: [Opensync-users] libsyncml internal error unknown response

[Ulrich B. <ubr...@t-...>]

Am Mittwoch, den 19.08.2009, 17:02 +0200 schrieb Michael Bell:

> I think it is possible to do implement this but first I have to
> understand why the message is not send.
> 

OK, you are right, a short Put sequence was send. but it was quite
different from the one in windows with the authentication enclosed.
I looked into the communication of libsyncml and my phone with strace
--ff.
I got the following sequence:
The bytes are in octal. 

write(10, "ATZ\r"..., 4)                = 4
select(11, [10], NULL, NULL, {1, 0})     = 1 (in [10], left {0, 852000})
read(10, "ATZ\r"..., 100)               = 4
select(11, [10], NULL, NULL, {1, 0})     = 1 (in [10], left {0, 996000})
read(10, "\r\nOK\r\n"..., 96)           = 6
write(10, "AT+CGMI\r"..., 8)            = 8
select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {0, 996000})
read(10, "AT+CGMI\r"..., 100)           = 8
select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {1, 0})
read(10, "\r\nSAMSUNG\r\n"..., 92)      = 11
select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {1, 0})
read(10, "\r\nOK\r\n"..., 81)           = 6
write(10, "AT+CGMM\r"..., 8)            = 8
select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {0, 996000})
read(10, "AT+CGMM\r"..., 100)           = 8
select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {0, 996000})
read(10, "\r\nE250i\r\n"..., 92)        = 9
select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {1, 0})
read(10, "\r\nOK\r\n"..., 83)           = 6
open("/usr/lib/gconv/gconv-modules.cache", O_RDONLY) = 11
fstat64(11, {st_mode=S_IFREG|0644, st_size=25700, ...}) = 0
mmap2(NULL, 25700, PROT_READ, MAP_SHARED, 11, 0) = 0xb7e07000
close(11)                               = 0
futex(0xb7b8ea6c, FUTEX_WAKE_PRIVATE, 2147483647) = 0
write(2, "\n** (process:5310): WARNING **: U"..., 96) = 96
write(10, "AT$TSSPCSW=1\r"..., 13)      = 13  
// I changed the normal AAT+CPROT=0 to this command, as it was used in
the windows software.
select(11, [10], NULL, NULL, {1, 0})     = 1 (in [10], left {0, 996000})
read(10, "AT$TSSPCSW=1\r"..., 100)      = 13
select(11, [10], NULL, NULL, {1, 0})    = 1 (in [10], left {1, 0})
read(10, "\r\nCONNECT\r\n"..., 87)      = 11
write(10, "\200\0\25\20\0\4\0F\0\16SYNCML-SYNC"..., 21) = 21
select(11, [10], NULL, NULL, {20, 0})     = 1 (in [10], left {19,
936000})
read(10, "\240\0\32\20\0\4\0\313\0\0\0\2J\0\16SYNCML-SYNC"..., 512) = 26

// OK you are right, here is a PUT sequence. The windows software
contains here the authentication sequence from the server.
write(10, "\202\0\337\313\0\0\0\2B\0 application/vnd.syncml"..., 223) =
223
select(11, [10], NULL, NULL, {20, 0})   = 1 (in [10], left {20, 0})
read(10, "\240\0\3"..., 512)            = 3
write(10, "\203\0(\313\0\0\0\2B\0 application/vnd.syncml"..., 40) = 40
select(11, [10], NULL, NULL, {20, 0})     = 1 (in [10], left {19,
952000})
// the phone E250i does not accept here. A phone E250 does not answer at
all at this place.
read(10, "\306\0\3"..., 512)            = 3
poll([{fd=5, events=POLLIN}], 1, 1)     = 0 (Timeout)
poll([{fd=5, events=POLLIN}], 1, 1) = 0 (Timeout)
poll([{fd=5, events=POLLIN}], 1, 1)     = 0 (Timeout)
poll([{fd=5, events=POLLIN}], 1, 1)     = 0 (Timeout)
poll([{fd=5, events=POLLIN}], 1, 1)     = 0 (Timeout)
poll([{fd=5, events=POLLIN}], 1, 1) = ? ERESTART_RESTARTBLOCK (To be
restarted)

Regards

Ulrich

Re: [Opensync-users] libsyncml internal error unknown response

[Michael B. <mic...@cm...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Ulrich,

can you please sniff the communication of libsyncml with the mobile
again? I would like to see what the library is really doing. The traces
in the Trac ticket show that there was a PUT request and you wrote there
was only a visible GET request.

from log:

smlTransportObexClientSend: Done sending the put request


This sounds really obscure. After the PUT request usually a GET request
follows. If the first PUT request happens and only the GET request
failed then we have "only" to modifiy the first SyncML message.

Best regards

Michael
- --
___________________________________________________________________

Michael Bell                        Humboldt-Universitaet zu Berlin

Tel.: +49 (0)30-2093 2482           ZE Computer- und Medienservice
Fax:  +49 (0)30-2093 2704           Unter den Linden 6
mic...@cm...       D-10099 Berlin
___________________________________________________________________

PGP Fingerprint: 09E4 3D29 4156 2774 0F2C  C643 D8BD 1918 2030 5AAB
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkqMFi0ACgkQ2L0ZGCAwWqv1GgCggQFSezgEHih8oNajawPNnx1Q
IOUAoOU3cVB5syhDUoJSKMPr7qfOCZ/c
=gx01
-----END PGP SIGNATURE-----

Re: [Opensync-users] libsyncml internal error unknown response

[Michael B. <mic...@cm...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ulrich Brammer wrote:
> Am Mittwoch, den 19.08.2009, 16:08 +0200 schrieb Michael Bell:
>>
>> I read your ticket and libsyncml sends a message to your mobile. You can
>> find it in Thread3079403264-4972.log of ticket #246.
>>
> I don't think that this message was really send. I tried to look into
> the communication with strace. I found only the obex get command from
> libsyncml. But the phone did not respond to this command. So the
> communication was not succesfull after the obex connection sequence.  

So the real problem is, why does libsyncml not send the message?

> I have no idea why in the windows software the server authenticates to
> the client. But this may be one of the bugs of the Teleca syncml stack.
> So is there a chance to handle this special behavior with libsyncml?

I think it is possible to do implement this but first I have to
understand why the message is not send.

Best regards

Michael
- --
___________________________________________________________________

Michael Bell                        Humboldt-Universitaet zu Berlin

Tel.: +49 (0)30-2093 2482           ZE Computer- und Medienservice
Fax:  +49 (0)30-2093 2704           Unter den Linden 6
mic...@cm...       D-10099 Berlin
___________________________________________________________________

PGP Fingerprint: 09E4 3D29 4156 2774 0F2C  C643 D8BD 1918 2030 5AAB
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkqMFBoACgkQ2L0ZGCAwWqsVowCgyelX6Qt7f3v3R2nuA5+LiteP
YngAn0ro9F6cvImvaypqiBVCe7jj10vg
=lk5q
-----END PGP SIGNATURE-----

Re: [Opensync-users] libsyncml internal error unknown response

[Ulrich B. <ubr...@t-...>]

Am Mittwoch, den 19.08.2009, 16:08 +0200 schrieb Michael Bell:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi Ulrich,
> 
> I read your ticket and libsyncml sends a message to your mobile. You can
> find it in Thread3079403264-4972.log of ticket #246.
> 
I don't think that this message was really send. I tried to look into
the communication with strace. I found only the obex get command from
libsyncml. But the phone did not respond to this command. So the
communication was not succesfull after the obex connection sequence.  

> Ulrich Brammer wrote:
> >    <LocURI>AT$CGSN
> > XXXXXX 01 XXXXXX 0</LocURI>
> 
> I have no idea what this is.
> 
> >   <Cred>
> >    <Meta>
> >     <Format xmlns="syncml:metinf">b64</Format>
> >     <Type xmlns="syncml:metinf">syncml:auth-basic</Type>
> >    </Meta>
> >    <Data>c2N0czpzY3Rz</Data>
> >   </Cred>
> 
> The OMA DS server authenticates itself to the mobile. Usually the scheme
> is that the client authenticates to the server.
> 
> BTW Data is a clear text passphrase.
> 
> The other stuff looks quite normal. So there are two issues to clarify:
> 
> 1. Which AT command is used as source?
> 2. Why does the server authenticates to the client?
> 
> Best regards
> 
> Michael
> - --
The Teleca Desktop Synchronisation Server seems to work the other way.
The Synchronisation Server first reads the IMEI or serial-mumber of the
phone with 
"AT$CGSN\r". This AT-command gives at my phone the same result as 
"AT+CGSN\r". The answer from the AT$CGSN\r is used as TargetURI (the
identification of the phone). 
I did replace some of the numbers of the IMEI with XXX to not show it to
everyone.

I have no idea why in the windows software the server authenticates to
the client. But this may be one of the bugs of the Teleca syncml stack.
So is there a chance to handle this special behavior with libsyncml?

Best regards

Ulrich