Menu
▾
▴
opensync-devel
|
From: deloptes <del...@ya...> - 2010-10-12 19:59:44
|
I'm not sure about this message do you have any idea developer@lisa:~/kde-testdir$ gdb syncml-ds-tool GNU gdb (GDB) 7.2-debian Copyright (C) 2010 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> ... Starting program: /opt/testing/opensync/bin/syncml-ds-tool -b xx:xx:xx:xx:xx:xx 7 [Thread debugging using libthread_db enabled] Program received signal SIGSEGV, Segmentation fault. 0x00007ffff5e23d8a in _IO_vfprintf_internal (s=0x7fffffffddc0, format=<value optimized out>, ap=0x7fffffffdf70) at vfprintf.c:1614 1614 vfprintf.c: No such file or directory. in vfprintf.c (gdb) |
|
From: Daniel G. <go...@b1...> - 2010-10-13 01:01:14
|
On Wednesday, October 13, 2010 08:59:11 am deloptes wrote: > developer@lisa:~/kde-testdir$ gdb syncml-ds-tool > GNU gdb (GDB) 7.2-debian > Copyright (C) 2010 Free Software Foundation, Inc. > License GPLv3+: GNU GPL version 3 or later > <http://gnu.org/licenses/gpl.html> > ... > Starting program: /opt/testing/opensync/bin/syncml-ds-tool -b > xx:xx:xx:xx:xx:xx 7 > [Thread debugging using libthread_db enabled] > > Program received signal SIGSEGV, Segmentation fault. > 0x00007ffff5e23d8a in _IO_vfprintf_internal (s=0x7fffffffddc0, > format=<value optimized out>, ap=0x7fffffffdf70) > at vfprintf.c:1614 > 1614 vfprintf.c: No such file or directory. > in vfprintf.c > (gdb) Could you send the full backtrace? Run: bt .. in gdb commandline. And make sure libsyncml and syncml-ds-tool are build with debuginfo - gcc flag -g ... Best Regards, Daniel -- Daniel Gollub Geschaeftsfuehrer: Ralph Dehner Linux Consultant & Developer Unternehmenssitz: Vohburg B1 Systems GmbH Amtsgericht: Ingolstadt Mobil: +49-(0)-160 47 73 970 Handelsregister: HRB 3537 EMail: go...@b1... http://www.b1-systems.de Adresse: B1 Systems GmbH, Osterfeldstraße 7, 85088 Vohburg http://pgpkeys.pca.dfn.de/pks/lookup?op=get&search=0xED14B95C2F8CA78D |
|
From: Emanoil K. <del...@ya...> - 2010-10-13 10:18:23
|
Thanks Daniel
--- On Wed, 10/13/10, Daniel Gollub <go...@b1...> wrote:
> From: Daniel Gollub <go...@b1...>
>
>
> Could you send the full backtrace?
>
> Run: bt .. in gdb commandline.
>
> And make sure libsyncml and syncml-ds-tool are build with
> debuginfo - gcc flag
> -g ...
>
I'm not sure I've understood exactly what you mean (or what I am supposed to do). I don't have that much experience or knowledge of the devs language.
I'm wondering if it is not better to wait until the weekend as then I'll be able to reboot the server and check again (I've installed some updates that affected glibc, but I couldn't reboot). I've also installed the latest kernel 2.26.35.7. I want to see if it will behave the same when everything is fresh.
Let me know if you think it is relevant.
I've also observed messages in dmesg when osplugin fails (I think it is comming from the threads)
Last but not least - I don't know how to make my mortal user talk to the BT device - it lacks privileges, or something else.
Can someone share config and experience with me please?!
regards
|
|
From: Chris F. <cd...@fo...> - 2010-10-14 23:03:56
|
On Wed, Oct 13, 2010 at 03:18:16AM -0700, Emanoil Kotsev wrote: > Thanks Daniel > > --- On Wed, 10/13/10, Daniel Gollub <go...@b1...> wrote: > > > From: Daniel Gollub <go...@b1...> > > ? ? ? ? ? ? ? ? > > > > Could you send the full backtrace? > > > > Run: bt .. in gdb commandline. > > > > And make sure libsyncml and syncml-ds-tool are build with > > debuginfo - gcc flag > > -g ... > > > > I'm not sure I've understood exactly what you mean (or what I am > supposed to do). I don't have that much experience or knowledge of the > devs language. There was a segfault reported in the original email, so he's asking for a gdb debugger stack trace. :-) ulimit -c unlimited <run program that crashes> gdb /some/program core > bt If the stack trace doesn't include function names, then you'll need to recompile the libraries with debug info enabled (gcc's -g flag). - Chris |
|
From: Emanoil K. <del...@ya...> - 2010-12-02 11:16:11
|
Hi,
I'm a bit disappointed, because no one is paying attentioin to what we are doing. There are many questions that can be answered in couple of minutes, but it takes weeks and months.
I've raized few tickets in libsyncml and libopensync because I finally managed to get the magic set of options for syncml-ds-tool to access Contacts, Calendar and Notes in the nokia 5530, which is S60 (not sure which revision).
Unfortunately Calendar does not work with syncml-ds-tool and Contacts break osynctool.
The wbxmls for Calendar are OK, but parser breaks (I think as I stated that this is because of DESCRIPTION field present).
Someone needs to look into this issue to help me improve the library.
This guy 'bellmich' seems to be dead. Could you kick his ass to be cooperative and get working the bugs out or let us discuss and seek someone who can take over.
The situation ATM is UNACCEPTABLE!
thanks in advance and regards
|
|
From: Chris F. <cd...@fo...> - 2010-12-02 13:57:32
|
Hi Emanoil, I've been busy with higher priority projects over the last months or so, and just now trying to finish off a few items pending on Barry. I'm about 122 messages behind on the opensync-devel mailing list, and that's not how I intended things to go. There'll be no "kicking of asses" from my end, though. :-) Everyone can only code as much as they are able. I hope to take a closer look at this today. Since I'm 122 messages behind, I'm quite out of the loop... if you can forward me your most important email or bug ticket that you would like help with, I can start from there. Thanks, - Chris On Thu, Dec 02, 2010 at 03:16:05AM -0800, Emanoil Kotsev wrote: > Hi, > > I'm a bit disappointed, because no one is paying attentioin to what we are doing. There are many questions that can be answered in couple of minutes, but it takes weeks and months. > > I've raized few tickets in libsyncml and libopensync because I finally managed to get the magic set of options for syncml-ds-tool to access Contacts, Calendar and Notes in the nokia 5530, which is S60 (not sure which revision). > > Unfortunately Calendar does not work with syncml-ds-tool and Contacts break osynctool. > > The wbxmls for Calendar are OK, but parser breaks (I think as I stated that this is because of DESCRIPTION field present). > > Someone needs to look into this issue to help me improve the library. > > This guy 'bellmich' seems to be dead. Could you kick his ass to be cooperative and get working the bugs out or let us discuss and seek someone who can take over. > > The situation ATM is UNACCEPTABLE! > > thanks in advance and regards > > > > |
|
From: Emanoil K. <del...@ya...> - 2010-12-02 14:17:00
|
Hi, Thanks for replying
--- On Thu, 12/2/10, Chris Frey <cd...@fo...> wrote:
>
> I've been busy with higher priority projects over the last
> months or so,
> and just now trying to finish off a few items pending on
> Barry.
> I'm about 122 messages behind on the opensync-devel mailing
> list, and
> that's not how I intended things to go.
>
> There'll be no "kicking of asses" from my end, though.
> :-) Everyone can
> only code as much as they are able.
Personal issues
*) That guy 'bellmich' - I think is Michael- has pending tickets for more then 8 months now that are not assessed. There has been no activity on the libsyncml code since a long time and the code is not working
*) No one is answering our questions in the dev list.
how are we supposed to help development if no docs and infos are provided.
The code issues
*) libsyncml is not working for Calendar (snycml v1.2) and
*) libopensync is not working for Contacts (seems that there are no issues with parsing the xml provided by syncml)
tickets on libsyncml side
#269 invalid utf8 passed to VFormat. Limping along. ( syncml issues ) Parser 0.5.4 libsyncml 0.6.0 defect blocker bellmich new 12/01/10
#270 invalid character found in attribute group/name: "41" String so far: ( syncml issues ) libsyncml 0.5.4 libsyncml 0.6.0 defect blocker bellmich new 12/01/10
#271 osync_converter_invoke: XMLFormat validation failed. ( syncml issues ) libsyncml 0.5.4 libsyncml 0.6.0 defect blocker bellmich new 12/01/10
#272 "DESCRIPTION:" in calendar items breaks parser libsyncml 0.5.4 libsyncml 0.6.0 defect blocker bellmich new 12/02/10
The syncml is definitely unstable and even worse than libopensync itself and tehn I've expected. We need rapid improvement here as nokia phones will probably fail syncing. I have 5530 with recent OS updated, but the issue is trivial - I guess may be change some xsd will fix it.
In opensync I have problem understanding ref/unref changes in akonadi-sync ... and no one answered what needs to be done and why other plugins have different implementation
I don't want to waste your time - I just need hints of people that know the api, so that I don't reinvent the wheel and spend time working on solving the issues then analysing and hacking the api.
It would be great if you let us focus on syncml together, so that I may (help to) debug and get some satisfying results before christmas this year.
For now akonadi-sync is doing really fine, so what I need is just to get the data out of the phone and even this is possible, but it gets stuck in libsyncml or libopensync.
thanks in advance and regards
regards
|
|
From: Michael B. <mic...@cm...> - 2010-12-12 20:27:20
Attachments:
smime.p7s
|
Hi, I actually write a bunch of mails to eplain my behaviour. Please accept my apologies for any inconvenience caused. Best regards Michael On 12/02/10 12:16, Emanoil Kotsev wrote: > Hi, > > I'm a bit disappointed, because no one is paying attentioin to what we are doing. There are many questions that can be answered in couple of minutes, but it takes weeks and months. > > I've raized few tickets in libsyncml and libopensync because I finally managed to get the magic set of options for syncml-ds-tool to access Contacts, Calendar and Notes in the nokia 5530, which is S60 (not sure which revision). > > Unfortunately Calendar does not work with syncml-ds-tool and Contacts break osynctool. > > The wbxmls for Calendar are OK, but parser breaks (I think as I stated that this is because of DESCRIPTION field present). > > Someone needs to look into this issue to help me improve the library. > > This guy 'bellmich' seems to be dead. Could you kick his ass to be cooperative and get working the bugs out or let us discuss and seek someone who can take over. > > The situation ATM is UNACCEPTABLE! > > thanks in advance and regards > > > > > > ------------------------------------------------------------------------------ > Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! > Tap into the largest installed PC base & get more eyes on your game by > optimizing for Intel(R) Graphics Technology. Get started today with the > Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. > http://p.sf.net/sfu/intelisp-dev2dev > _______________________________________________ > Opensync-devel mailing list > Ope...@li... > https://lists.sourceforge.net/lists/listinfo/opensync-devel -- ___________________________________________________________________ Michael Bell Humboldt-Universitaet zu Berlin Tel.: +49 (0)30-2093 70143 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 |
|
From: Emanoil K. <del...@ya...> - 2010-10-13 16:47:41
|
Another question for you that we may discuss until the weekend. As I mentioned I'm using evo2-sync as a pair for testing akonadi-sync and I've noticed that evo2-sync is reporting notes in vnote11 format and not journal (which is breaking the sync as akonadi-sync expects vjournal), which makes me think it is picking up the first available format from the opposite member. It is either the plugin or (what I think) the libopensync itself, selecting the first available path for the "note" sync. How can I tell opensync (from akonadi POV) I can and I want vjournal. I've tried setting preferred format in discover but it does not help much. The odd thing is that evolution does not report vnote11 as supported at all. Do I need to file bugs ... in normal life I would. here a portion to give you an idea akonadi_opensync(29750) DataSink::commit: change uid: "20100910T224808Z-11504-1010-1-0@lisa" akonadi_opensync(29750) DataSink::commit: objform: vnote11 akonadi_opensync(29750) DataSink::commit: akonadi_opensync(29750) DataSink::commit: data "BEGIN:VNOTE VERSION:1.1 CATEGORIES:first CREATED:20100910T224808Z LAST-MODIFIED:20100910T224808Z SUMMARY:asdfasdf UID:20100910T224808Z-11504-1010-1-0@lisa END:VNOTE " akonadi_opensync(29750) DataSink::collection: akonadi_opensync(29750) DataSink::setPayload: akonadi_opensync(29750) DataSink::setPayload: notes ERROR: Broken Pipe ERROR: Broken Pipe ERROR: Broken Pipe ERROR: Trying to send to a queue thats not connected ERROR: Broken Pipe OK! Completing synchronization! ERROR: Broken Pipe ERROR: Broken Pipe ERROR: Broken Pipe Error for entry 20100910T224808Z-11504-1010-1-0@lisa (vnote11) and member 2 (akonadi-sync): Broken Pipe Mapping Error: Broken Pipe EXIT_ERROR: osync_obj_engine_commit_change_callback: Broken Pipe and this data is coming from evo2. and evo2 is reporting vjournal. Hm ... from the way I see how evolution works I don't think I'll upgrade my current debian system soon. Perhaps it would be better to get the syncing with kde3 done for now. However I would like to work on this plugin to get it stable until I move to kde4. I'm also not sure if I have to try fixing evo2-sync the way I think it should work and proposing a patch. It would be very nice if someone knowing evo api gets involved. regards --- On Wed, 10/13/10, Daniel Gollub <go...@b1...> wrote: > From: Daniel Gollub <go...@b1...> > Subject: Re: [Opensync-devel] syncml-ds-tool > To: ope...@li..., del...@ya... > Date: Wednesday, October 13, 2010, 1:41 AM > On Wednesday, October 13, 2010 > 08:59:11 am deloptes wrote: > > developer@lisa:~/kde-testdir$ gdb syncml-ds-tool > > GNU gdb (GDB) 7.2-debian > > Copyright (C) 2010 Free Software Foundation, Inc. > > License GPLv3+: GNU GPL version 3 or later > > <http://gnu.org/licenses/gpl.html> > > ... > > Starting program: > /opt/testing/opensync/bin/syncml-ds-tool -b > > xx:xx:xx:xx:xx:xx 7 > > [Thread debugging using libthread_db enabled] > > > > Program received signal SIGSEGV, Segmentation fault. > > 0x00007ffff5e23d8a in _IO_vfprintf_internal > (s=0x7fffffffddc0, > > format=<value optimized out>, > ap=0x7fffffffdf70) > > at vfprintf.c:1614 > > 1614 vfprintf.c: No such file or > directory. > > in vfprintf.c > > (gdb) > > > Could you send the full backtrace? > > Run: bt .. in gdb commandline. > > And make sure libsyncml and syncml-ds-tool are build with > debuginfo - gcc flag > -g ... > > Best Regards, > Daniel > > -- > Daniel Gollub > Geschaeftsfuehrer: > Ralph Dehner > Linux Consultant & Developer > Unternehmenssitz: Vohburg > B1 Systems GmbH > Amtsgericht: > Ingolstadt > Mobil: +49-(0)-160 47 73 970 > Handelsregister: HRB > 3537 > EMail: go...@b1... > http://www.b1-systems.de > > Adresse: B1 Systems GmbH, Osterfeldstraße 7, 85088 > Vohburg > http://pgpkeys.pca.dfn.de/pks/lookup?op=get&search=0xED14B95C2F8CA78D > |
|
From: deloptes <del...@ya...> - 2010-10-15 20:30:59
|
Chris Frey wrote: > If the stack trace doesn't include function names, then you'll need to > recompile the libraries with debug info enabled (gcc's -g flag). Last question - how do I do this in cmake everything I've tried shows no debug symbols in gdb the new thing this time was this message warning: Can't read pathname for load map: Input-/Outputerror. but it is just a warning thanks in advance |
|
From: Chris F. <cd...@fo...> - 2011-01-14 20:49:35
|
On Fri, Oct 15, 2010 at 10:30:21PM +0200, deloptes wrote: > Chris Frey wrote: > > > If the stack trace doesn't include function names, then you'll need to > > recompile the libraries with debug info enabled (gcc's -g flag). > > Last question - how do I do this in cmake > > everything I've tried shows no debug symbols in gdb Sorry for the late reply, but I encourage people to setup their cmake like I suggested in the Plugin Porting Howto from a while back, or at least use it as a base for their own config: http://opensync.org/wiki/devel/pluginPortingGuide-0.40 Cmake seems to hide things by default, which, for developers, is bad. - Chris |
|
From: deloptes <del...@ya...> - 2010-10-13 20:01:42
|
It seems usb is working better
syncml-ds-tool -u
Superuser privileges are required to access complete USB information.
Found 2 USB OBEX interfaces
Interface 0:
Manufacturer: Nokia
Product: Nokia 5530 XpressMusic
Interface description: SYNCML-SYNC
Interface 1:
Manufacturer: Nokia
Product: Nokia 5530 XpressMusic
Interface description: PC Suite Service
--discover works well
but sync does not
Main sink of member 2 of type akonadi-sync just connected
contact sink of member 2 of type akonadi-sync just connected
Main sink of member 1 of type syncml-obex-client just connected
ERROR: Forbidden (0x43)
EXIT_ERROR: _recv_event: Forbidden (0x43)
ERROR: Forbidden (0x43)
contact sink of member 1 of type syncml-obex-client had an error: Forbidden
(0x43)
further testing syncml-ds-tool with gdb I see
Reading symbols from /opt/testing/opensync/bin/syncml-ds-tool...(no
debugging symbols found)...done.
(gdb) r -u
Starting program: /opt/testing/opensync/bin/syncml-ds-tool -u
[Thread debugging using libthread_db enabled]
Superuser privileges are required to access complete USB information.
Found 2 USB OBEX interfaces
Interface 0:
Manufacturer: Nokia
Product: Nokia 5530 XpressMusic
Interface description: SYNCML-SYNC
Interface 1:
Manufacturer: Nokia
Product: Nokia 5530 XpressMusic
Interface description: PC Suite Services
Use '-u interface_number' to connect
** Message: syncml-ds-tool failed.
Program exited with code 01.
(gdb) r -u 0
Starting program: /opt/testing/opensync/bin/syncml-ds-tool -u 0
[Thread debugging using libthread_db enabled]
Program received signal SIGSEGV, Segmentation fault.
0x00007ffff5e29d8a in _IO_vfprintf_internal (s=0x7fffffffde40, format=<value
optimized out>, ap=0x7fffffffdff0) at vfprintf.c:1614
1614 vfprintf.c: Datei oder Verzeichnis nicht gefunden.
in vfprintf.c
I then configured the phone properly (I believe)
and got some meaningful output though the tool failed again
however if I try osynctool with --sync I get
ERROR: Forbidden (0x43)
EXIT_ERROR: _recv_event: Forbidden (0x43)
ERROR: Forbidden (0x43)
contact sink of member 1 of type syncml-obex-client had an error: Forbidden
(0x43)
ERROR: At least one sink_engine failed while connecting
ERROR: No objtypes left without error. Aborting
ERROR: No objtypes left without error. Aborting
What can be this?
Anyway I've got tired for today, please give me some meaningful feedback.
Thanks and regards
|
|
From: Michael B. <mb...@gm...> - 2010-10-14 16:15:19
|
On Wed, Oct 13, 2010 at 10:01:15PM +0200, deloptes wrote: > EXIT_ERROR: _recv_event: Forbidden (0x43) As far as I know, Forbidden (0x43) is a rather generic error message. Try changing the version of syncml (e.g. 1.1 vs. 1.2) you request, if it's a Nokia mobile, try adding "PC Suite" (or whatever the name was) as identifier, and possibly try to turn on/off wbxml. I think there is a wiki page on libsyncml.opensnyc.org somewhere which discusses this. Michael |
|
From: deloptes <del...@ya...> - 2010-10-14 16:30:20
|
Michael Banck wrote: > On Wed, Oct 13, 2010 at 10:01:15PM +0200, deloptes wrote: >> EXIT_ERROR: _recv_event: Forbidden (0x43) > > As far as I know, Forbidden (0x43) is a rather generic error message. thanks > > Try changing the version of syncml (e.g. 1.1 vs. 1.2) you request, if > it's a Nokia mobile, try adding "PC Suite" (or whatever the name was) as > identifier, and possibly try to turn on/off wbxml. I did the version change ... to match the one on the phone syncml-ds-tool started but broke in the middle. May be I have to apply your next suggestion > > I think there is a wiki page on libsyncml.opensnyc.org somewhere which > discusses this. > Where - the one that explains how to configure the svn plugin? Are you using USB or Bluetooth? and Why there is only "Bluetooth/Internet" in the Sync configure menu. Are there docs on it? Thanks |
Thanks for helping keep SourceForge clean.
X