[OpenSIPS-Devel] New SIP SIMPLE client SDK release 0.12.0
Adrian Georgescu writes: > The software is available as a tar archive, darcs repository or as a > Debian package for Debian unstable distribution. since python-sipsimple depends on python-eventlet, i tried to build the latter by getting the source with 'apt-get source python-eventlet' and then building debian package with command 'dpkg-buildpackage -rfakeroot -b -us -uc'. the latter failed as shown below. -- juha ... cd . && python setup.py install --root=/usr/src/eventlet-0.8.10/debian/python-eventlet/ --no-compile -O0 running install running build running build_py running install_lib creating /usr/src/eventlet-0.8.10/debian/python-eventlet/usr creating /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib creating /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5 creating /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages creating /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet copying build/lib/eventlet/httpdate.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet copying build/lib/eventlet/wsgi.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet creating /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/support copying build/lib/eventlet/support/pycurls.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/support copying build/lib/eventlet/support/greenlets.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/support copying build/lib/eventlet/support/stacklesss.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/support copying build/lib/eventlet/support/pylib.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/support copying build/lib/eventlet/support/stacklesspypys.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/support copying build/lib/eventlet/support/__init__.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/support copying build/lib/eventlet/support/twisteds.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/support copying build/lib/eventlet/corolocal.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet copying build/lib/eventlet/api.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet copying build/lib/eventlet/proc.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet copying build/lib/eventlet/timer.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet copying build/lib/eventlet/logutil.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet copying build/lib/eventlet/saranwrap.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet creating /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/hubs copying build/lib/eventlet/hubs/libev.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/hubs copying build/lib/eventlet/hubs/twistedr.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/hubs copying build/lib/eventlet/hubs/__init__.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/hubs copying build/lib/eventlet/hubs/selects.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/hubs copying build/lib/eventlet/hubs/nginx.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/hubs copying build/lib/eventlet/hubs/hub.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/hubs copying build/lib/eventlet/hubs/poll.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/hubs copying build/lib/eventlet/hubs/libevent.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet/hubs copying build/lib/eventlet/__init__.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet copying build/lib/eventlet/processes.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet copying build/lib/eventlet/backdoor.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet copying build/lib/eventlet/jsonhttp.py -> /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/lib/python2.5/site-packages/eventlet creating /usr/src/eventlet-0.8.10/debian/python-eventlet/usr/li
Re: [OpenSIPS-Devel] New SIP SIMPLE client SDK release 0.12.0
Adrian Georgescu writes: > I have pushed the python-eventlet 0.8.11 to our repository can you try > again now? adrian, looks like apt-get source still gives 0.8.10 version: apt-get update ... Get:11 http://ag-projects.com unstable Release [5267B] Ign http://ag-projects.com unstable Release Ign http://ag-projects.com unstable/main Sources j...@rautu:/usr/src$ apt-get source python-eventlet Reading package lists... Done Building dependency tree Reading state information... Done Need to get 177kB of source archives. Get:1 http://ag-projects.com unstable/main eventlet 0.8.10 (dsc) [830B] Get:2 http://ag-projects.com unstable/main eventlet 0.8.10 (tar) [176kB] but that is ok. i can pull the source using darcs. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] New SIP SIMPLE client SDK release 0.12.0
adrian,
i just darcs'ed python-eventlet:
darcs get http://devel.ag-projects.com/repositories/python-eventlet
which gave me version 0.8.11:
/usr/src/python-eventlet$ more debian/changelog
python-eventlet (0.8.11) unstable; urgency=low
* Renamed package to python-eventlet
-- Adrian Georgescu Sun, 13 Dec 2009 15:30:54 +0100
...
but still, building of debian package fails.
-- juha
---
j...@rautu:/usr/src/python-eventlet$ dpkg-buildpackage -rfakeroot -b -us -uc
dpkg-buildpackage: set CFLAGS to default value: -g -O2
dpkg-buildpackage: set CPPFLAGS to default value:
dpkg-buildpackage: set LDFLAGS to default value:
dpkg-buildpackage: set FFLAGS to default value: -g -O2
dpkg-buildpackage: set CXXFLAGS to default value: -g -O2
dpkg-buildpackage: source package python-eventlet
dpkg-buildpackage: source version 0.8.11
dpkg-buildpackage: source changed by Adrian Georgescu
dpkg-buildpackage: host architecture i386
...
running install_egg_info
Traceback (most recent call last):
File "setup.py", line 30, in
"Development Status :: 4 - Beta"]
File "/usr/lib/python2.5/distutils/core.py", line 151, in setup
dist.run_commands()
File "/usr/lib/python2.5/distutils/dist.py", line 974, in run_commands
self.run_command(cmd)
File "/usr/lib/python2.5/distutils/dist.py", line 994, in run_command
cmd_obj.run()
File "/usr/lib/python2.5/site-packages/setuptools/command/install.py", line
53, in run
return _install.run(self)
File "/usr/lib/python2.5/distutils/command/install.py", line 510, in run
self.run_command(cmd_name)
File "/usr/lib/python2.5/distutils/cmd.py", line 333, in run_command
self.distribution.run_command(command)
File "/usr/lib/python2.5/distutils/dist.py", line 993, in run_command
cmd_obj.ensure_finalized()
File "/usr/lib/python2.5/distutils/cmd.py", line 117, in ensure_finalized
self.finalize_options()
File
"/usr/lib/python2.5/site-packages/setuptools/command/install_egg_info.py", line
22, in finalize_options
self.set_undefined_options('install',('install_layout','install_layout'))
File "/usr/lib/python2.5/distutils/cmd.py", line 309, in set_undefined_options
getattr(src_cmd_obj, src_option))
File "/usr/lib/python2.5/distutils/cmd.py", line 112, in __getattr__
raise AttributeError, attr
AttributeError: install_layout
make: *** [python-install-py] Error 1
___
Devel mailing list
Devel@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] New SIP SIMPLE client SDK release 0.12.0
Saúl Ibarra Corretgé writes:
> I just tried and I'm getting the correct version both for i386 and amd64
> can you do apt-get clean && apt-get autoclean && apt-get update &&
> apt-get source python-eventlet and check if it still occurs?
saul,
i'm able to get version source 0.8.11 both from your deb archive and
from darcs archive, but in both cases, building of debian package fails
like shown below.
perhaps some build dependency is missing from debian/control?
-- juha
--
running install_egg_info
Traceback (most recent call last):
File "setup.py", line 30, in
"Development Status :: 4 - Beta"]
File "/usr/lib/python2.5/distutils/core.py", line 151, in setup
dist.run_commands()
File "/usr/lib/python2.5/distutils/dist.py", line 974, in run_commands
self.run_command(cmd)
File "/usr/lib/python2.5/distutils/dist.py", line 994, in run_command
cmd_obj.run()
File "/usr/lib/python2.5/site-packages/setuptools/command/install.py", line
53, in run
return _install.run(self)
File "/usr/lib/python2.5/distutils/command/install.py", line 510, in run
self.run_command(cmd_name)
File "/usr/lib/python2.5/distutils/cmd.py", line 333, in run_command
self.distribution.run_command(command)
File "/usr/lib/python2.5/distutils/dist.py", line 993, in run_command
cmd_obj.ensure_finalized()
File "/usr/lib/python2.5/distutils/cmd.py", line 117, in ensure_finalized
self.finalize_options()
File
"/usr/lib/python2.5/site-packages/setuptools/command/install_egg_info.py", line
22, in finalize_options
self.set_undefined_options('install',('install_layout','install_layout'))
File "/usr/lib/python2.5/distutils/cmd.py", line 309, in set_undefined_options
getattr(src_cmd_obj, src_option))
File "/usr/lib/python2.5/distutils/cmd.py", line 112, in __getattr__
raise AttributeError, attr
AttributeError: install_layout
make: *** [python-install-py] Error 1
___
Devel mailing list
Devel@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] openxcap does not start with python-application 1.2.2
latest openxcap version 1.1.2 starts fine in daemon mode IF python-application is version 1.1.5: # /etc/init.d/openxcap start Starting OpenXCAP server: openxcap Feb 16 04:49:05 localhost openxcap[8866]: Log opened. Feb 16 04:49:05 localhost openxcap[8866]: Starting OpenXCAP 1.1.2 . Feb 16 04:49:06 localhost openxcap[8866]: Trusted peers: 127.0.0.1 Feb 16 04:49:06 localhost openxcap[8866]: xcap.server.HTTPFactory starting on 8080 Feb 16 04:49:06 localhost openxcap[8866]: 5 xcap documents in the database if i upgrade python-application to latest version 1.2.2, openxcap does not anymore start in daemon mode: dpkg -i python-application_1.2.2_all.deb (Reading database ... 87152 files and directories currently installed.) Preparing to replace python-application 1.1.5 (using python-application_1.2.2_all.deb) ... Unpacking replacement python-application ... Setting up python-application (1.2.2) ... Processing triggers for python-support ... r...@rautu:/usr/src# /etc/init.d/openxcap stop Stopping OpenXCAP server: openxcap . r...@rautu:/usr/src# Feb 16 04:50:37 localhost openxcap[8866]: Received SIGTERM, shutting down. Feb 16 04:50:37 localhost openxcap[8866]: (Port 8080 Closed) Feb 16 04:50:37 localhost openxcap[8866]: Main loop terminated. r...@rautu:/usr/src# /etc/init.d/openxcap start Starting OpenXCAP server: openxcap . and does not report anything to syslog. it does start in no-fork mode: r...@rautu:/usr/src# openxcap --no-fork Starting OpenXCAP 1.1.2 Trusted peers: 127.0.0.1 xcap.server.HTTPFactory starting on 8080 5 xcap documents in the database -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] New SIP SIMPLE client SDK release 0.12.0
Saúl Ibarra Corretgé writes: > Which Debian version and python-setuptools package are you using? I'll > try to reproduce the problem. saul, i'm using debian lenny with some newer python packages from debian testing. setuptools version is ii python-setuptools 0.6.10-1 -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] openxcap does not start with python-application 1.2.2
Saúl Ibarra Corretgé writes: > You're right, OpenXCAP needs to be adjusted in order to work well with > latest python-application. good to know that i didn't screw up something. > For the moment, as you guessed, using python-application version 1.1.5 > is the solution. When I fix this I'll come back to you so you can > upgrade. i found this out when i tried to use python-sipsimple depending on python-application (>= 1.2.1) on the same host as openxcap. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] New SIP SIMPLE client SDK release 0.12.0
Saúl Ibarra Corretgé writes: > I just did a fresh Debian Lenny install and it worked for me, so now I'm > confused :-S saul, thanks for your help. debian lenny without newer versions of some python packages and some new python packages does not allow running openxcap. i'll try adding these openxcap required python packages one at a time to see at which point python-eventlet build starts to fail. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] New MediaProxy release 2.4.0 with support for ICE negotiation
Adrian Georgescu writes: > mediaproxy (2.4.0) unstable; urgency=low > >* Added ICE negotiation capability and assert itself as a TURN relay > candidate in the SDP, for a complete description and usage > information > see http://mediaproxy-ng.org/wiki/ICE >* Fix race in which an already called event tried to be cancelled >* Prevent other acounting backends to be affected by a failure in > one of > them adrian, the list does not include anything about python-application 1.2.2 compatibility. does it work with that version or still require old version? -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] New MediaProxy release 2.4.0 with support for ICE negotiation
Juha Heinanen writes: > adrian, > > the list does not include anything about python-application > 1.2.2 compatibility. does it work with that version or still require > old version? > > -- juha sorry about the mixup. the problem was with openxcap, not mediaproxy. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] openxcap does not start with python-application 1.2.2
any news about the subject? debian squeeze has python-application version 1.2.3, which makes it impossible to run openxcap on that system. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] openxcap does not start with python-application 1.2.2
Adrian Georgescu writes: > We work on packing as we speak. No ETA as usual :-) that is great news, -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] New OpenXCAP release 1.2.1
Adrian Georgescu writes: > There is a new release of OpenXCAP, version 1.2.1. It contains bug > fixes. adrian, i built and installed this on debian squeeze. it went fine except this warning during startup: Starting OpenXCAP server: openxcap /usr/lib/python2.6/dist-packages/zope/__init__.py:3: UserWarning: Module _mysql was already imported from /usr/lib/pymodules/python2.6/_mysql.so, but /usr/lib/pymodules/python2.6 is being added to sys.path import pkg_resources looks like it is harmless, but would anyway be nice to get rid off. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] New OpenXCAP release 1.2.1
Adrian Georgescu writes: > This is harmless and is caused by some other non-xcap lib loading a > previously loaded module. Not sure how we can rid of it because is not > written by us . thanks for confirmation. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] [ opensips-Bugs-3036174 ] presence/timer related crash
> check the mailing list for reply, also see rev 7069 on 1.6 branch. bogdan, thanks for the fix. somehow i managed to miss the reply on the users list. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] Proposal for adding OMA extensions support for pres-rules
- (presence_xml) When running in integrated mode OpenSIPS doesn't know the xcap-root, so it's not able to determine if a reference is external or not. Thus, I propose xcap_server parameter to be used also in combination with integrated mode, this way OpenSIPS can know if a reference to a resource-list is internal or external. OpenSIPS will ignore external references in integrated mode. saul, why can't you have it both ways, i.e., follow external references also in integrated mode? -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] opensips 1.6 is still crashing at presence timer
bogdan, you wrote a couple of weeks ago that the presence timer related crash that i reported has been fixed. today i tried with latest 1.6 and still got the crash: Program terminated with signal 11, Segmentation fault. #0 0xb75c53e3 in db_mysql_val2bind (v=0xbfd85e5c, binds=0x8188078, i=1) at val.c:274 274 *(binds[i].is_null) = 0; (gdb) where #0 0xb75c53e3 in db_mysql_val2bind (v=0xbfd85e5c, binds=0x8188078, i=1) at val.c:274 #1 0xb75c093e in db_mysql_do_prepared_query (conn=, v=, n=2, uv=0x0, un=0, query=) at dbase.c:443 #2 0xb75c20dd in db_mysql_delete (_h=0x817c1a0, _k=0xbfd85e90, _o=0xbfd85e84, _v=0xbfd85e48, _n=2) at dbase.c:893 #3 0xb7274ea4 in msg_watchers_clean (ticks=3600, param=0x0) at subscribe.c:484 #4 0x080b64cf in timer_ticker () at timer.c:325 #5 run_timer_process () at timer.c:395 #6 start_timer_processes () at timer.c:475 #7 0x08069ea3 in main_loop (argc=16, argv=0xbfd86054) at main.c:867 #8 main (argc=16, argv=0xbfd86054) at main.c:1388 -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] trying to examine presence/timer crash
i tried to figure out why i'm getting the presence/timer crashes, but
don't know how to access the argument value of a function.
(gdb) where
#0 0xb758e3e3 in db_mysql_val2bind (v=0xbfcbdf3c, binds=0x8188190, i=1)
at val.c:274
#1 0xb758993e in db_mysql_do_prepared_query (conn=,
v=, n=2, uv=0x0, un=0, query=)
at dbase.c:443
#2 0xb758b0dd in db_mysql_delete (_h=0x817c1a0, _k=0xbfcbdf70,
_o=0xbfcbdf64, _v=0xbfcbdf28, _n=2) at dbase.c:893
#3 0xb723dea4 in msg_watchers_clean (ticks=3600, param=0x0)
at subscribe.c:484
#4 0x080b64cf in timer_ticker () at timer.c:325
#5 run_timer_process () at timer.c:395
#6 start_timer_processes () at timer.c:475
#7 0x08069ea3 in main_loop (argc=16, argv=0xbfcbe134) at main.c:867
#8 main (argc=16, argv=0xbfcbe134) at main.c:1388
i would like know what struct db_val_t* argument v of of
db_mysql_do_prepared_query function contains:
static int db_mysql_do_prepared_query(const db_con_t* conn, const str *query,
const db_val_t* v, int n, const db_val_t* uv, int un)
i go to frame 1
(gdb) frame 1
#1 0xb758993e in db_mysql_do_prepared_query (conn=,
v=, n=2, uv=0x0, un=0, query=)
at dbase.c:443
443 if (db_mysql_val2bind( v+i , mysql_bind, i)<0 ) {
and check args:
(gdb) info args
conn =
v =
n = 2
uv = 0x0
un = 0
query =
(gdb) whatis v
type = const db_val_t *
but when i try to print type field of first v value, i get an error:
(gdb) print v->type
Cannot access memory at address 0x0
same for other fields of v.
what is it that i'm doing wrong?
-- juha
___
Devel mailing list
Devel@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] trying to examine presence/timer crash
Juha Heinanen writes:
> (gdb) where
> #0 0xb758e3e3 in db_mysql_val2bind (v=0xbfcbdf3c, binds=0x8188190, i=1)
> at val.c:274
> #1 0xb758993e in db_mysql_do_prepared_query (conn=,
> v=, n=2, uv=0x0, un=0, query=)
> at dbase.c:443
> #2 0xb758b0dd in db_mysql_delete (_h=0x817c1a0, _k=0xbfcbdf70,
> _o=0xbfcbdf64, _v=0xbfcbdf28, _n=2) at dbase.c:893
> #3 0xb723dea4 in msg_watchers_clean (ticks=3600, param=0x0)
> at subscribe.c:484
> #4 0x080b64cf in timer_ticker () at timer.c:325
> #5 run_timer_process () at timer.c:395
> #6 start_timer_processes () at timer.c:475
> #7 0x08069ea3 in main_loop (argc=16, argv=0xbfcbe134) at main.c:867
> #8 main (argc=16, argv=0xbfcbe134) at main.c:1388
a bit more about this. at frame 2, i can get value of v. the function
and call of db_mysql_do_prepared_query:
int db_mysql_delete(const db_con_t* _h, const db_key_t* _k, const db_op_t* _o,
const db_val_t* _v, const int _n)
...
ret = db_mysql_do_prepared_query(_h, &query_holder, _v, _n, NULL, 0);
(gdb) frame 2
#2 0xb758b0dd in db_mysql_delete (_h=0x817c1a0, _k=0xbfcbdf70,
_o=0xbfcbdf64, _v=0xbfcbdf28, _n=2) at dbase.c:893
893 ret = db_mysql_do_prepared_query(_h, &query_holder, _v,
_n, NULL, 0);
(gdb) info args
_h = 0x817c1a0
_k = 0xbfcbdf70
_o = 0xbfcbdf64
_v = 0xbfcbdf28
_n = 2
(gdb) print _v->type
$7 = DB_INT
but when in frame 1 (in db_mysql_do_prepared_query):
static int db_mysql_do_prepared_query(const db_con_t* conn, const str *query,
const db_val_t* v, int n, const db_val_t* uv, int un)
...
if (db_mysql_val2bind( v+i , mysql_bind, i)<0 ) {
(gdb) info args
conn =
v =
n = 2
uv = 0x0
un = 0
query =
gdb) print v->type
Cannot access memory at address 0x0
perhaps v has changed in db_mysql_do_prepared_query before
db_mysql_val2bind call?
-- juha
___
Devel mailing list
Devel@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] trying to examine presence/timer crash
Bogdan-Andrei Iancu writes: > in frame 1, try to print: > > p (struct prep_stmt*)(*(conn)->curr_ps) > > p ((struct prep_stmt*)(*(conn)->curr_ps))->ctx same thing, won't work: (gdb) p (struct prep_stmt*)(*(conn)->curr_ps) Cannot access memory at address 0x4 (gdb) p ((struct prep_stmt*)(*(conn)->curr_ps))->ctx Cannot access memory at address 0x4 -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] trying to examine presence/timer crash
Bogdan-Andrei Iancu writes: > So I should understand that conn->curr_ps is 4 ? This is strange as the > bt shows the processing going further where this *conn->curr_ps is > used.maybe the "4" value is a side-effect of the original bug.. > > if I make a small patch to print some logs at runtime (to try to hunt > done this issue), would it be possible to run it ? sure, -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] [OpenSIPS-Users] New OpenXCAP release 2.0.0 with support for OMA XDM SIP SIMPLE specifications
Iñaki Baz Castillo writes: > So such PUT will be rejected, but the existing document (containing > external references) will remain *active*. perhaps you can write an external script so remove the external references when you change configuration of xcap server. changing the config file is an external action too with regards to the running server. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] contact param missing in active_watchers table
i noticed that when presence server receives subscribe that has this kind of contact header: Contact: sip:u...@172.17.49.100:5060;alias=x then ;alias param in NOT included in active_watchers contact field. but if presence server receives this kind of contact header: Contact: then ;alias param is correctly included in active_watchers contact field. not including the param if <>s are missing looks like the bug to me, since in both cases ;alias should be interpreted as uri parameter, not header parameter. comments? -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] contact param missing in active_watchers table (resend)
i don't know what happened, when i never got this back from the list server. it is in the archives though. my mail logs show only one message today from devel list and it was from anca. original contents is below. - i noticed that when presence server receives subscribe that has this kind of contact header: Contact: sip:u...@172.17.49.100:5060;param=x then ;param in NOT included in active_watchers contact field and notify is not delivered properly. but if presence server receives this kind of contact header: Contact: then ;param param is correctly included in active_watchers contact field. not including the param if <>s are missing around the uri looks like a bug to me, since according to sip syntax in both cases ;param should be interpreted as uri parameter, not header parameter. comments? -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] [ opensips-Feature Requests-3115782 ] presence: unique record update for PUBLISH w/o SIP-If-Match
> Let's discus the MWI case. If a PUBLISH is received by the presence > server, a record is created and a 200ok with 'SIP-ETag' header is sent > back. > If a new PUBLISH is received and the PUBLISH does not have a > 'SIP-If-Match' header that will match the previous 'SIP-ETag' header, > a new record will be created for the same subscription. Instead of > creating a new record, the existing record should be updated: the > state of the voicemail box is unique per subscriber. > This approach is not compliant with the rfc 3903, but it make sense to > implement the above behavior for certain event types packages like > mwi. ovidiu, my mwi application first checks if etag exists and depending on the result, either uses existing one or "." in the new publish. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] [ opensips-Feature Requests-3115782 ] presence: unique record update for PUBLISH w/o SIP-If-Match
Ovidiu Sas writes: > I was thinking about extending the sever behavior for this particular > kind of event packages for which a single record in the presentity > table is enough to maintain the event state. that too would be fine with me. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] compilation warning with stable 1.6 release
i compiled today latest 1.6 from svn and got this kind of compilation warning: make[2]: Entering directory `/usr/src/5.1-src/openxg-pres-serv/modules/presence_xcapdiff' Compiling presence_xcapdiff.c presence_xcapdiff.c: In function ‘mod_init’: presence_xcapdiff.c:106: warning: passing argument 4 of ‘pua.add_event’ from incompatible pointer type presence_xcapdiff.c:106: note: expected ‘int (*)(struct publ_info *, struct str **, int, struct str *)’ but argument is of type ‘int (*)(struct publ_info_t *, struct str **, int, struct str **)’ it looks scary to me in a stable release. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] SF.net SVN: opensips:[7502] trunk/modules/presence/subscribe.c
Anca Vamanu writes: > Log Message: > --- > - fixed crash in prepare statements (thanks to Duane Larson for help with > debugging) > > Modified Paths: > -- > trunk/modules/presence/subscribe.c anca, hopefully this is the same that i have reported. once you backport to 1.6, i'll give it a try. i got the crash also today. -- juha gdb) where #0 0xb74ec3e3 in db_mysql_val2bind (v=0xbfcd7a9c, binds=0x81990e0, i=1) at val.c:274 #1 0xb74e793e in db_mysql_do_prepared_query (conn=, v=, n=2, uv=0x0, un=0, query=) at dbase.c:443 #2 0xb74e90dd in db_mysql_delete (_h=0x818d158, _k=0xbfcd7ad0, _o=0xbfcd7ac4, _v=0xbfcd7a88, _n=2) at dbase.c:893 #3 0xb71af094 in msg_watchers_clean (ticks=3600, param=0x0) at subscribe.c:484 #4 0x080b8e1f in timer_ticker () at timer.c:325 #5 run_timer_process () at timer.c:395 #6 start_timer_processes () at timer.c:475 #7 0x0806c32c in main_loop (argc=16, argv=0xbfcd7c94) at main.c:867 #8 main (argc=16, argv=0xbfcd7c94) at main.c:1388 ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] SF.net SVN: opensips:[7490] branches/1.6/modules/presence
Ovidiu Sas writes: > - applied patch #3058434 from Kennard White - a module parameter that > - enables including closed status body in Notify when no published > - info is found for the user ovidiu, does the above apply only to presence events or all events? it does not seem to make sense, for example, for message-summary events. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] SF.net SVN: opensips:[7502] trunk/modules/presence/subscribe.c
Anca Vamanu writes: > Yes, it's the same. I suppose you also had rls module loaded? Because > this is when I think it showed. yes, i have rls module loaded. if i don't see the crash anymore in a few days, i'll close the ticket. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] presentity table version change
i tried to run new version of presence module that would include anca's crash fix. the result was that my presence server didn't start: Dec 10 20:19:01 sip /usr/sbin/pres-serv[20005]: ERROR:core:db_check_table_version: invalid version 4 for table presentity found, expected 5 i consider it a very bad idea to introduce database version change in the middle of a stable release, which may be result of this: presence: support for extra headers in NOTIFY - extra headers received in PUBLISH requests can be stored and sent to the subscrber via NOTIFY requests now getting a serious bug fixed means that also database table needs to be upgraded for a reason that has nothing to do with fixing bugs. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] presentity table version change
Ovidiu Sas writes: > The new feature was in the trunk for a while. > The opensips 1.6.4 release is scheduled for mid December. The feature > was backported from trunk in order to prepare the release. in my opinion a new release should happen by copying trunk to a new branch. before that happens only bug fixes should go to current stable release and all new features should go to trunk. now there is now way one can get the crash fixed except by updating database because of some new feature that should only exist in trunk until the release actually happens. stable branches should not be playgrounds for new features, especially ones that are not backwards compatible. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] presentity table version change
the new field in presentity table has been defined like this: extra_hdrs BLOB NOT NULL how can i upgrade an existing table when the field cannot be null and there is no default value? -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] presentity table version change
Ovidiu Sas writes: > Fixed in revision 7505. ovidiu, thanks and sorry about my rant yesterday. i was tired after long flights. but i still feel that it would be cleaner to have real 1.6.x branches in svn rather than just 1.6 branch that now includes both bug fixes and backwards incompatible new features. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] presentity table version change
Ovidiu Sas writes: > Probably 1.6.4 will still be release from the existing 1.6 branch (as > many backports are already present in the 1.6 branch). > Post 1.6.4 releases (eg 1.6.5+) will have their own branch. ok, thanks, -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] tls/tls_server.c does not compile
i tried to build my presence server from latest svn trunk and got this: Compiling tls/tls_server.c tls/tls_server.c: In function ‘tls_blocking_write’: tls/tls_server.c:744: error: expected expression before ‘int’ tls/tls_server.c:744: error: expected ‘)’ before ‘;’ token make[1]: *** [tls/tls_server.o] Error 1 -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] empty publish results in invalid notify body
i did some presence tests with sip-communicator and twinkle. when i
tell sip-communicator that i'm offline, it sends a publish with empty
body like this:
PUBLISH sip:t...@test.fi SIP/2.0.
Call-ID: a915618a37062a26f07f35ba7d538...@0.0.0.0.
CSeq: 9 PUBLISH.
From: "Test" ;tag=50de411e.
To: "Test" .
Max-Forwards: 69.
Content-Type: application/pidf+xml.
Contact: "Test" .
User-Agent: SIP Communicator1.0-alpha6-nightly.build.3174Linux.
Expires: 0.
Event: presence.
SIP-If-Match: a.1293545574.14775.16.0.
Via: SIP/2.0/UDP x.x.x.x;branch=z9hG4bK7d28.92f1bd43.0.
Via: SIP/2.0/UDP
192.168.0.169:11417;rport=63441;received=x.x.x.x;branch=z9hG4bK-383134-a1b11a5839f6b262b0926731752edbfb.
Content-Length: 0.
once presence server gets it, it sends out a notify like this:
NOTIFY sip:j...@192.168.0.169:5074;alias=x.x.x.x~63373~2;transport=tcp SIP/2.0.
Via: SIP/2.0/UDP x.x.x.x;branch=z9hG4bK4eec.58bca9d.0.
Via: SIP/2.0/UDP 127.0.0.1:5082;branch=z9hG4bK4eec.ebd886e3.0.
To: ;tag=lgwpg.
From: ;tag=48de7c3cb9f0a00e953b10ae4e49e9de-0f9f.
CSeq: 10 NOTIFY.
Call-ID: vctlhxdlljfi...@sip.test.fi.
Route:
,.
Content-Length: 144.
User-Agent: OpenSIPS (1.6.4-tls (i386/linux)).
Max-Forwards: 69.
Event: presence.
Contact: .
Subscription-State: active;expires=132.
Content-Type: application/pidf+xml.
.
closed
when twinkle receives the notify, it claims that it has invalid sip
body.
indeed, rfc3863 in section 4.1.1 says:
The element MUST contain a namespace declaration ('xmlns')
to indicate the namespace on which the presence document is based.
The presence document compliant to this specification MUST have the
namespace 'urn:ietf:params:xml:ns:pidf:'.
looks like namespace declaration is missing from the notify document that
presence server has generated on its own as response to empty publish body.
should i open a ticket on this or have i misunderstood something?
-- juha
___
Devel mailing list
Devel@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] empty publish results in invalid notify body
Juha Heinanen writes: > > id="t5374">closed i tried to read presentity.c code and if the above doc is created by build_offline_presence function, it tried to add the missing stuff to the doc: xmlNewProp(root_node, BAD_CAST "xmlns", BAD_CAST "urn:ietf:params:xml:ns:pidf"); xmlNewProp(root_node, BAD_CAST "xmlns:dm", BAD_CAST "urn:ietf:params:xml:ns:pidf:data-model"); xmlNewProp(root_node, BAD_CAST "xmlns:rpid", BAD_CAST "urn:ietf:params:xml:ns:pidf:rpid" ); xmlNewProp(root_node, BAD_CAST "xmlns:c", BAD_CAST "urn:ietf:params:xml:ns:pidf:cipid"); so far i have not been able to figure out, why nothing happens. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] new kind of presence crash
today i noticed two new kind of presence crashes like the one below. -- juha (gdb) where #0 0xb75f8425 in ?? () from /lib/i686/cmov/libc.so.6 #1 0xb711694b in xmlOutputBufferWriteString () from /usr/lib/libxml2.so.2 #2 0xb71c3fa8 in ?? () from /usr/lib/libxml2.so.2 #3 0xb71c4881 in ?? () from /usr/lib/libxml2.so.2 #4 0xb71c40ad in ?? () from /usr/lib/libxml2.so.2 #5 0xb71c4881 in ?? () from /usr/lib/libxml2.so.2 #6 0xb71c40ad in ?? () from /usr/lib/libxml2.so.2 #7 0xb71c3b61 in ?? () from /usr/lib/libxml2.so.2 #8 0xb71c5053 in xmlDocDumpFormatMemoryEnc () from /usr/lib/libxml2.so.2 #9 0xb71c51df in xmlDocDumpMemory () from /usr/lib/libxml2.so.2 #10 0xb70bb92a in agregate_xmls (pres_user=0xbf8c08b0, pres_domain=0xbf8c08c0, body_array=0x81a7dbc, n=2) at notify_body.c:621 #11 0xb70bc447 in pres_agg_nbody (pres_user=0xbf8c08b0, pres_domain=0xbf8c08c0, body_array=0x81a7dbc, n=2, off_index=-1) at notify_body.c:76 #12 0xb7202f44 in get_p_notify_body (pres_uri=..., event=0xb31dd598, etag=0x0, contact=0xbf8c0ce0, dbody=0x0, extra_hdrs=0xbf8c0b1c) at notify.c:982 #13 0xb7207c14 in build_str_hdr (subs=0xbf8c0c88, is_body=0, hdr=0x0, extra_hdrs=0x0) at notify.c:146 #14 0xb7208c1f in printf_subs (subs=0x0, watcher_subs=0x0, n_body=0x0, force_null_body=0, extra_hdrs=0x0) at notify.c:102 #15 send_notify_request (subs=0x0, watcher_subs=0x0, n_body=0x0, force_null_body=0, extra_hdrs=0x0) at notify.c:1844 #16 0xb721d14c in update_subscription (msg=0x81a6220, subs=0xbf8c0c88, init_req=0) at subscribe.c:389 #17 0xb7220785 in handle_subscribe (msg=0x81a6220, force_active_param=0x0, str2=0x0) at subscribe.c:685 #18 0x08058279 in do_action (a=0x81a013c, msg=0x81a6220) at action.c:1195 #19 0x08056c2c in run_action_list (a=0x81a013c, msg=0x81a6220) at action.c:140 #20 0x0809ade1 in eval_elem (e=0x81a01a8, msg=0x81a6220, val=0x0) at route.c:1354 #21 eval_expr (e=0x81a01a8, msg=0x81a6220, val=0x0) at route.c:1696 #22 0x0809a991 in eval_expr (e=0x81a01d4, msg=0x81a6220, val=0x0) at route.c:1712 #23 0x0809a9d9 in eval_expr (e=0x81a0200, msg=0x81a6220, val=0x0) at route.c:1717 #24 0x08058508 in do_action (a=0x81a03b0, msg=0x81a6220) at action.c:802 #25 0x08056c2c in run_action_list (a=0x819f6bc, msg=0x81a6220) at action.c:140 #26 0x0805a2cf in do_action (a=0x81a0488, msg=0x81a6220) at action.c:819 #27 0x08056c2c in run_action_list (a=0x819eef8, msg=0x81a6220) at action.c:140 #28 0x0805a06a in run_actions (a=0x819ad4c, msg=0x81a6220) at action.c:120 #29 do_action (a=0x819ad4c, msg=0x81a6220) at action.c:488 #30 0x08056c2c in run_action_list (a=0x819ad4c, msg=0x81a6220) at action.c:140 #31 0x0805a2cf in do_action (a=0x819ae24, msg=0x81a6220) at action.c:819 #32 0x08056c2c in run_action_list (a=0x819ae24, msg=0x81a6220) at action.c:140 #33 0x0805ba09 in run_actions (a=0x819ae24, msg=0x81a6220) at action.c:120 #34 run_top_route (a=0x819ae24, msg=0x81a6220) at action.c:181 #35 0x0808fe99 in receive_msg ( buf=0x8172300 "SUBSCRIBE sip:127.0.0.1:5082 SIP/2.0\r\nVia: SIP/2.0/UDP x.x.x.x;branch=z9hG4bKa677.33ea0a24.0\r\nVia: SIP/2.0/UDP x.x.x.x;branch=z9hG4bKa677.137a5be5.0;i=d\r\nVia: SIP/2.0/TCP 192.168.0.169:5074"..., len=655, rcv_info=0xbf8c1f58) at receive.c:162 #36 0x080c9ce0 in udp_rcv_loop () at udp_server.c:492 #37 0x0806d2e3 in main_loop (argc=16, argv=0xbf8c2104) at main.c:824 #38 main (argc=16, argv=0xbf8c2104) at main.c:1393 ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] empty publish results in invalid notify body
Juha Heinanen writes: > i tried to read presentity.c code and if the above doc is created by > build_offline_presence function, it tried to add the missing stuff to > the doc: perhaps the bug is in presence_xml/notify_body.c instead. i'll make some experiments. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] empty publish results in invalid notify body
Juha Heinanen writes: > perhaps the bug is in presence_xml/notify_body.c instead. i'll make > some experiments. here is diff that works for me. could someone check and commit it. -- juha *** /usr/src/orig/opensips/branches/1.6/modules/presence_xml/notify_body.c 2010-12-28 15:48:35.0 +0200 --- notify_body.c 2010-12-31 11:51:32.0 +0200 *** *** 713,718 --- 713,721 root_node->ns = NULL; } + xmlNewProp(root_node, BAD_CAST "xmlns", + BAD_CAST "urn:ietf:params:xml:ns:pidf"); + xmlDocSetRootElement(new_doc, root_node); tuple_node= xmlCopyNode(tuple_node, 2); ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] new kind of presence crash
Anca Vamanu writes: > I noticed this crash also. I think it is due to the recently added > xmlFreeNsList function when copying a node. I removed this and I didn't > observe the crash anymore. I have committed the fix on svn. Please > update and let me know if you still see the crash. anca, thanks for the fix. i'll update a couple of servers to the latest version and watch for possible crashes. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] New SylkServer release 1.1.0
adrian, thanks for the new version of sylkserver. i built it myself on debian squeeze by first getting it from your repo with command apt-get source sylkserver build and install went fine, but when i try to start sylkserver, i get the error below. any idea what goes wrong? my python-sipsimple version is 0.18.0. -- juha Mar 24 11:39:45 sip sylk-server[3613]: Starting SylkServer 1.1.0, config=/etc/sylkserver/config.ini Mar 24 11:39:47 sip sylk-server[3613]: using set_wakeup_fd Mar 24 11:39:48 sip sylk-server[3613]: fatal error: failed to create SylkServer: /usr/lib/pymodules/python2.6/sipsimple/core/_core.so: undefined symbol: pjsip_msg_find_remove_hdr_by_name Mar 24 11:39:48 sip sylk-server[3613]: Traceback (most recent call last): Mar 24 11:39:48 sip sylk-server[3613]: File "/usr/bin/sylk-server", line 86, in main Mar 24 11:39:48 sip sylk-server[3613]: from sylk.server import SylkServer Mar 24 11:39:48 sip sylk-server[3613]: File "/usr/lib/pymodules/python2.6/sylk/server.py", line 11, in Mar 24 11:39:48 sip sylk-server[3613]: from sipsimple.account import Account, BonjourAccount, AccountManager Mar 24 11:39:48 sip sylk-server[3613]: File "/usr/lib/pymodules/python2.6/sipsimple/account.py", line 32, in Mar 24 11:39:48 sip sylk-server[3613]: from sipsimple.core import ContactHeader, Credentials, Engine, FromHeader, FrozenSIPURI, Registration, RouteHeader, SIPURI, Subscription, ToHeader, PJSIPError, SIPCoreError Mar 24 11:39:48 sip sylk-server[3613]: File "/usr/lib/pymodules/python2.6/sipsimple/core/__init__.py", line 4, in Mar 24 11:39:48 sip sylk-server[3613]: from sipsimple.core._core import * Mar 24 11:39:48 sip sylk-server[3613]: ImportError: /usr/lib/pymodules/python2.6/sipsimple/core/_core.so: undefined symbol: pjsip_msg_find_remove_hdr_by_name ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] common-policy.xsd not found after openxcap upgrade
i upgraded openxcap on debian lenny from version 1.0.6 to 1.0.7 using .deb package. after upgrade, openxcap server does not start, because common-policy.xsd file is not found: Starting OpenXCAP 1.0.6 Supported Root URIs: http://xcap.tutpro.com/xcap-root Traceback (most recent call last): File "/usr/bin/openxcap", line 55, in from xcap.server import XCAPServer File "/usr/lib/python2.5/site-packages/xcap/server.py", line 17, in File "/usr/lib/python2.5/site-packages/xcap/authentication.py", line 22, in File "/usr/lib/python2.5/site-packages/xcap/appusage/__init__.py", line 523, in File "/usr/lib/python2.5/site-packages/xcap/appusage/__init__.py", line 68, in __init__ IOError: [Errno 2] No such file or directory: '/usr/lib/python2.5/site-packages/xcap/appusage/../xml-schemas/common-policy.xsd' the link is there, but it points to wrong location: /usr/lib/python2.5/site-packages/xcap/xml-schemas# ls -ls total 0 0 lrwxrwxrwx 1 root root 54 2008-09-09 17:18 common-policy.xsd -> /usr/share/pyshared/xcap/xml-schemas/common-policy.xsd openxcap package installs the file in # dpkg -L openxcap | egrep common-policy.xsd /usr/share/python-support/openxcap/xcap/xml-schemas/common-policy.xsd -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] rls_handle_subscribe() return code when list is not found
from rls tutorial document:
http://www.opensips.org/Resources/Rls
i get impression that rls_handle_subscribe() should return
to_presence_code, if resource list document is not found in xcap server.
however, when i read the code, it has:
if(doc== NULL|| service_node==NULL)
{
LM_DBG("list not found - search for uri =
%.*s\n",subs.pres_uri.len,
subs.pres_uri.s);
reply_code = 404;
reply_str = pu_404_rpl;
goto error;
}
is this a bug? if not, i don't see how a normal handle_subscribe()
can be tried if rls document doesn't exit.
-- juha
___
Devel mailing list
Devel@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] rls_handle_subscribe() return code when list is not found
Iñaki Baz Castillo writes: > I haven't inspected the code, but rls_handle_subscribe() does what > it's supposed to do (return the preconfigured to $rc when no RL doc is > found). inaki, are you sure that it returns pre-configured $rc even when Supported: header includes 'eventlist', but there is no presence list document in xcap server? if there is no 'eventlist' in Supported header, then pre-configured $rc is returned as it should. my understand from reading the code and from doing some tests is that in the above mentioned combination things, -1 is incorrectly returned. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] rls_handle_subscribe() return code when list is not found
Anca Vamanu writes: > I have now looked through the RFC and to be honest, I think that I > shouldn't have done the change and still let the Subscribe go to the > presence server. I don't remember which was the reason to change it at > that moment, but I remember it came after a discussion with Dan Pascu > when he gave me a strong argument to do so. I am willing to change it > back, but will first wait for Dan's input on this. anca, i noticed this problem with real commercial sip ua that sends subscribe with eventlist in supported header even when subscribe is not for a list. in fact, i don't see any chance how ua would know beforehand if target of subscription is a list or not. in supported header it just tells that it can handle eventlist. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] SF.net SVN: opensips:[6041] branches/1.5/modules/rls/subscribe.c
Anca Vamanu writes: > - revert change made in commit rev #4944: return to_presence code for > - Subscribe messages that contain Supported:eventlist header but for > - which a list definition is not found anca, in case of in-dialog subscribe, the module still returns 404 if document is not found. isn't the idea that rls_handle_subscribe() can be called also for in-dialog requests and get 'to_presence_code' back if the in-dialog subscribe was not for rls server? -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] SF.net SVN: opensips:[6041] branches/1.5/modules/rls/subscribe.c
anca, regarding in-dialog subscribe requests, perhaps it could be checked in the script if the request is for rls server or not, for example, checking that request uri matches server_address. if so, then rls_handle_subscribe would not need to handle non-list subscribes. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] SF.net SVN: opensips:[6043] branches/1.5/modules/rls/subscribe.c
Anca Vamanu writes: > - if an indialog Subscribe does not match a known dialog, send the > - Subscribe to presence server (reported by Juha Heinanen) anca, thanks for the fix. after thinking a bit more this, it should be possible to distinguish in script if an initial subscribe possibly is for rls server, because normal subscribes are for uris of local users. so if target of subscribe is not for a local user, handle it with rls_handle_subscribe and return 404 if it fails. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] SF.net SVN: opensips:[6043] branches/1.5/modules/rls/subscribe.c
Adrian Georgescu writes: > You can subscribe also to users of remote domains and they are not > local. sure, but in that case, i'm not calling handle_subscribe, but just t_relay the request. > > so if target of > > subscribe is not for a local user, handle it with rls_handle_subscribe > > and return 404 if it fails. > > This is wrong for the reason above. sorry, i meant, user part of uri does not belong to existing user, but host part is a local domain. > There was a reason for handling the way it was, too bad Dan is in > holiday and could not comment. the way it was does not work if you plan to call handle_subscribe after rls_handle_subscribe. there is two possible implementations of rls_handle_subscribe: - do not do any smarts in the implementation, but leave it to the script to figure out when to call rls_subscribe - do smarts the way they are done now they way the smarts were done earlier, was not correct. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] rls_handle_notify() does not recognize dialog
i did some rls tests. first i create a buddy list sip:jh-budd...@test.fi
that currently has only one buddy t...@test.fi.
then jh used sip_subscribe_rls to subscribe to jh-buddies. wireshark
shows how rls server upon receiving subscribe send subscribe to
sip:t...@test.fi, how presence server replies to it with 200 ok followed
by in-dialog notify. when rls_handle_notify() handles this notify, it
returns 2 meaning that "the Notify did not belog to a dialog initiated
by the RLS server.".
why is that? the packets are below starting from subscribe sent by rls
server. "404 not found" comes from this piece of config code:
$var(ret_code) = rls_handle_notify();
if ($var(ret_code) == 1) exit;
if ($var(ret_code) == 2) {
sl_send_reply("404", "Not found");
exit;
};
-- juha
-
No. TimeSourceDestination Protocol
Info
9 15:53:46.955023 127.0.0.1 127.0.0.1 SIP
Request: SUBSCRIBE sip:sip:t...@test.fi
Frame 9 (448 bytes on wire, 448 bytes captured)
Linux cooked capture
Internet Protocol, Src: 127.0.0.1 (127.0.0.1), Dst: 127.0.0.1 (127.0.0.1)
User Datagram Protocol, Src Port: qcp (5082), Dst Port: qcp (5082)
Session Initiation Protocol
Request-Line: SUBSCRIBE sip:sip:t...@test.fi SIP/2.0
Message Header
Via: SIP/2.0/UDP 127.0.0.1:5082;branch=z9hG4bK8d99.0df60be7.0
To: sip:sip:t...@test.fi
From: ;tag=374ba85b6c1b7e45caf839d18042c8ae-ec2b
CSeq: 10 SUBSCRIBE
Call-ID: 7bffc7ef-21...@127.0.0.1
Content-Length: 0
User-Agent: OpenSIPS (1.5.2-notls (i386/linux))
Max-Forwards: 70
Event: presence
Contact:
Expires: 610
No. TimeSourceDestination Protocol
Info
10 15:53:46.956119 127.0.0.1 127.0.0.1 SIP
Status: 202 OK
Frame 10 (423 bytes on wire, 423 bytes captured)
Linux cooked capture
Internet Protocol, Src: 127.0.0.1 (127.0.0.1), Dst: 127.0.0.1 (127.0.0.1)
User Datagram Protocol, Src Port: qcp (5082), Dst Port: qcp (5082)
Session Initiation Protocol
Status-Line: SIP/2.0 202 OK
Message Header
Via: SIP/2.0/UDP 127.0.0.1:5082;branch=z9hG4bK8d99.0df60be7.0
To: sip:sip:t...@test.fi;tag=48de7c3cb9f0a00e953b10ae4e49e9de-3f5d
From: ;tag=374ba85b6c1b7e45caf839d18042c8ae-ec2b
CSeq: 10 SUBSCRIBE
Call-ID: 7bffc7ef-21...@127.0.0.1
Expires: 610
Contact:
Server: OpenSIPS (1.5.2-notls (i386/linux))
Content-Length: 0
No. TimeSourceDestination Protocol
Info
11 15:53:46.956273 127.0.0.1 127.0.0.1 SIP
Request: NOTIFY sip:r...@127.0.0.1:5082
Frame 11 (505 bytes on wire, 505 bytes captured)
Linux cooked capture
Internet Protocol, Src: 127.0.0.1 (127.0.0.1), Dst: 127.0.0.1 (127.0.0.1)
User Datagram Protocol, Src Port: qcp (5082), Dst Port: qcp (5082)
Session Initiation Protocol
Request-Line: NOTIFY sip:r...@127.0.0.1:5082 SIP/2.0
Message Header
Via: SIP/2.0/UDP 127.0.0.1:5082;branch=z9hG4bK0f26.4e8e9573.0
To: ;tag=374ba85b6c1b7e45caf839d18042c8ae-ec2b
From: ;tag=48de7c3cb9f0a00e953b10ae4e49e9de-3f5d
CSeq: 1 NOTIFY
Call-ID: 7bffc7ef-21...@127.0.0.1
Content-Length: 0
User-Agent: OpenSIPS (1.5.2-notls (i386/linux))
Max-Forwards: 70
Event: presence
Contact:
Subscription-State: pending;expires=610
No. TimeSourceDestination Protocol
Info
12 15:53:46.956465 127.0.0.1 127.0.0.1 SIP
Status: 404 Not found
Frame 12 (378 bytes on wire, 378 bytes captured)
Linux cooked capture
Internet Protocol, Src: 127.0.0.1 (127.0.0.1), Dst: 127.0.0.1 (127.0.0.1)
User Datagram Protocol, Src Port: qcp (5082), Dst Port: qcp (5082)
Session Initiation Protocol
Status-Line: SIP/2.0 404 Not found
Message Header
Via: SIP/2.0/UDP 127.0.0.1:5082;branch=z9hG4bK0f26.4e8e9573.0
To: ;tag=374ba85b6c1b7e45caf839d18042c8ae-ec2b
From: ;tag=48de7c3cb9f0a00e953b10ae4e49e9de-3f5d
CSeq: 1 NOTIFY
Call-ID: 7bffc7ef-21...@127.0.0.1
Server: OpenSIPS (1.5.2-notls (i386/linux))
Content-Length: 0
___
Devel mailing list
Devel@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] rls_handle_notify() does not recognize dialog
Anca Vamanu writes: > I have just tested myself and did not saw this problem. First, what > version are you running? Trunk or 1.5 branch? i did my test with version 1.5. > Then, do you see this problem all the time or only from time to time? i have tried several times and always rls_handle_notify() fails to match the notify to the subscribe it sent. > I am thinking it might be a synchronization issue. And the processing of > the Notify taking place before the process that received the 200 OK for > Subscribe succeeds to update the dialog information. this is an idle presence server running on the same host as client. wireshark shows that notify comes before 1 ms has passed from 200 ok. > If you see this > problem only time to time, then this theory could be true and to prove > it you could try to run opensips with fork=no. If this problem doesn't > occur anymore then I was right. And then I will think at a solution.. I > have one in mind already. i'll try to try with fork=no. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] rls_handle_notify() does not recognize dialog
Anca Vamanu writes: > If you see this problem only time to time, then this theory could be > true and to prove > it you could try to run opensips with fork=no. If this problem doesn't > occur anymore then I was right. And then I will think at a solution.. I > have one in mind already. > I will wait for your answers. anca, i tried with fork=no with the same result, i.e., rls_handle_notify() returns for the notify. i saved the pcap trace in case you need it. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] rls_handle_notify() does not recognize dialog
anca, i think i found the problem related to rls subscription. rls sip:jh-budd...@test.fi contains: the uri is wrong, i.e., it contains double "sip:". also pua table has "sip:sip:t...@test.fi" in pres_uri column. i'll try to fix jh-buddies list. on the other hand, it might be a good idea to check in rls server that resource list uris are valid. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] pua_publish mi command fails
has something changed, because i noticed that now with 1.5 and
trunk pua_publish mi command fails. according to debug, everything goes
fine until 200 ok is received by pua module that then calls
mi_publ_rpl_cback with NULL parameter:
Sep 11 15:16:18 localhost /usr/sbin/pres-serv[3963]: DBG:pua:publ_cback_func:
completed with status 200 [contact:sip:j...@vm.test.fi]
Sep 11 15:16:18 localhost /usr/sbin/pres-serv[3963]: DBG:pua:publ_cback_func:
update record
Sep 11 15:16:18 localhost /usr/sbin/pres-serv[3963]: DBG:pua:run_pua_callbacks:
found callback
Sep 11 15:16:18 localhost /usr/sbin/pres-serv[3963]:
ERROR:pua_mi:mi_publ_rpl_cback: NULL parameter
according to the source, the test giving the error message is this:
if(reply== NULL || hentity== NULL || hentity->cb_param== NULL)
{
LM_ERR("NULL parameter\n");
return -1;
}
how come that a good 200 ok response to publish calls mi callback with
null parameter(s)?
-- juha
___
Devel mailing list
Devel@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] pua_publish mi command fails
Anca Vamanu writes: > I have fixed the error. It was indeed introduced recently when fixing > some other bug :). I will not port it to 1.5 branch yet because I have > noticed a problem when running with pua_usrloc and I want to fix that > also before porting. anca, i tried today with latest trunk version if pua_publish would work in it, but it turned out that neither subscribwes or publish requests work, instead, they both (as well as notify) give error: Sep 19 19:28:25 lohi /usr/sbin/pres-serv[8821]: WARNING:core:run_actions: null action list (rec_level=1) -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] pua_publish mi command fails
Bogdan-Andrei Iancu writes:
> If you do have a route[0] {} in your script, please replace it with
> route {}.
>
> Due latest changes on the named route, it might be a backward
> compatibility issue.
bogdan,
yes, i had route[0] { ... } in the script and run_actions warnings went
away, when i replaced it with route { ... }.
-- juha
___
Devel mailing list
Devel@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] publish problem with trunk
after upgrading from 1.5 to trunk, i have started to get these kind of messages to syslog: Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: INFO: Handling PUBLISH Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: ERROR:presence_xml:get_final_notify_body: while parsing the xml body message Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: ERROR:presence_xml:pres_apply_auth: in function get_final_notify_body Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: ERROR:presence:send_notify_request: in function apply_auth_nbody Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: ERROR:presence:notify: sending Notify not successful Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: ERROR:presence:publ_notify: Could not send notify for presence the publish request is normal refresh publish without body. -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] publish problem with trunk
Juha Heinanen writes: > after upgrading from 1.5 to trunk, i have started to get these kind of > messages to syslog: > > Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: INFO: Handling PUBLISH > > Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: > ERROR:presence_xml:get_final_notify_body: while parsing the xml body message > Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: > ERROR:presence_xml:pres_apply_auth: in function get_final_notify_body > Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: > ERROR:presence:send_notify_request: in function apply_auth_nbody > Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: ERROR:presence:notify: > sending Notify not successful > Sep 24 10:54:15 localhost /usr/sbin/pres-serv[26313]: > ERROR:presence:publ_notify: Could not send notify for presence > > the publish request is normal refresh publish without body. here is how the xml doc of t...@test.fi looks like in presentity table: \nopen\n -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
Re: [OpenSIPS-Devel] publish problem with trunk
Anca Vamanu writes: > I have tested exactly this - a refresh Publish with no body and > everything worked all right. I have added in that error message to print > the body that it attempts to parse - maybe this will give more info. > Please update from svn. anca, i built new package from svn and now i don't anymore see those error messages. i don't know what happened, but the problem has disappeared. thanks, juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] buggy event_parser in parse_event.c?
i started to wonder why i'm sometimes seeing these kind of error messages in syslog while starting opensips trunk presence server: Oct 4 11:38:02 localhost /usr/sbin/pres-serv[11306]: NOTICE:presence:mod_init: initializing module ... Oct 4 11:38:02 localhost /usr/sbin/pres-serv[11306]: ERROR:core:parse_params: invalid character, ; expected Oct 4 11:38:02 localhost /usr/sbin/pres-serv[11306]: ERROR:presence:contains_event: parsing event 'presence.winfo' (i added stuff to the last error message to see the event whose parsing fails.) then i started to read parse_event.c/event_parser() code and it didn't make sense to me: - after parsing the event, it goes to check if next char is ';': if( (*end)== ';') is that correct, because there may not be a next char, i.e., end may point to random byte after the event string? if it does so, error results if the next char happens to be ';'. - why is event sting copied to buf, for which i don't see any use in the function? -- juha ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
[OpenSIPS-Devel] buggy event_parser in parse_event.c?
Juha Heinanen writes: > then i started to read parse_event.c/event_parser() code and it didn't > make sense to me: > > - after parsing the event, it goes to check if next char is ';': > > if( (*end)== ';') > > is that correct, because there may not be a next char, i.e., end may > point to random byte after the event string? if it does so, error > results if the next char happens to be ';'. > > - why is event sting copied to buf, for which i don't see any use in the > function? i'm proposing the enclosed diff to parse_event.c in order to get rid of invalid error messages and unused buf. -- juha parse_event.c-diff Description: parse_event.c-diff ___ Devel mailing list Devel@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
