Hello,
(cross-posting as this decision may be done properly with feedback from
both developers and users community).
Now that we have some band new modules but also several that were not
packaged by Kamailio, just by ser (due to single modules directory), I
was thinking we may want to reconsider a bit the packages we do for v4.0.0.
1) A new package is related to the bunch of IMS modules, that is pretty
much obvious.
2) There are few modules targeting development samples or
troubleshooting: print, print_lib and malloc_test - I think they should
not be packaged.
3) I would create a special package for the modules using uid-based
database schema: db2_ops uid_auth_db uid_avp_db uid_domain uid_gflags
uid_uri_db uri_db
4) Maybe we should merge some packages by dependencies:
- a) some depend on libxml2: cpl-c xhttp_pi xmlrpc xmlops
- b) some depend on libssl: tls auth_identity outbound
5) Some experimental modules with external deps will stay unpackaged
like so far: db_oracle
Opinions?
Cheers,
Daniel
--
Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Kamailio World Conference, April 16-17, 2013, Berlin
- http://conference.kamailio.com -
Hi, I read in http://kamailio.org/docs/modules/1.4.x/nathelper.html that
Kamailio might rewrite ip address in o= in SDP to facilitate NAT traversal.
It may decide to use to original IP or use IP of RTP server. What factors
affect its decision ?
--
Khoa Pham
HCMC University of Science
Faculty of Information Technology
Hello,
We are using SQL Server 2008 as our database. When the sip_capture module tries to insert a row, the following error is given:
db_unixodbc [con.c:220]: unixodbc:SQLExecDirect=42000:1:156:[FreeTDS][SQL Server]Incorrect syntax near the keyword 'authorization'.
Using Kamailio 3.3.4.
Is there a way to force brackets around the column values?
Regards,
Grant
Hi,
this last weekend we went live from the old 3.0 version to the latest stable release of Kamailio (3,3,0). All went well, except for one issue. We make use of uac_replace_from in our config, in case a client needs CLIR: uac_replace_from("Anonymous","sip:anonymous@anonymous.invalid");
Since the upgrade we receive the following error:
uac [replace.c:521]: new URI shorter than old URI
Which causes the call to be disconnected after about 20/30 seconds. This, because the ACK is received by Kamailio, but cannot be 'linked' to the transaction i guess.
What is exactly causing this issue and, more importantly, is there any way to resolve this?
Thanks in advance,
Regards,
Ronald Voermans
Hello,
Can someone tell me which is executed first ?
the on reply route setup with t_on_reply() or the callbacks in the code
that allows the module dialog to insert the confirmed dialog into the
dialog table ?
Thanks,
David
Hello thereI hope that you can help me, I have an issue installing Kamailio, when I issue th command "make all" I receive the following error resolve.h:471:4: error: #error neither gethostbyname2 or getipnodebyname present
make: *** [action.o] Error 1
I am using debian squeeze, please let me know if there is something missing in my system or what is not allowing to install the SIP Server THanks Julián
Hi folks,
we have a serious issue with tm module. In Fail route we would like to
send a reply 404 when a cr_route function fails(no destination found).
We use t_reply("404", "No route to destination"). However after the
call is disconnected with a 404 reply the tm module writes a following
warning
WARNING: tm [t_lookup.c:1543]: WARNING: script writer didn't release transaction
Very soon after this error message is shown a segfault occurs. Log is
as follows:
Mar 3 15:26:46 hostname kamailio[791]: ERROR: carrierroute
[cr_func.c:594]: desired routing domain doesn't exist, prefix
0039333333333, carrier 14, domain 2
Mar 3 15:26:46 hostname kamailio[791]: WARNING: tm [t_lookup.c:1543]:
WARNING: script writer didn't release transaction
//ACK is received
Mar 3 15:26:52 hostname kernel: [771134.093250] kamailio[796]:
segfault at 4 ip 0815c182 sp bf8ace00 error 6 in
kamailio[8048000+1ea000]
Mar 3 15:26:53 hostname kamailio[803]: : <core> [pass_fd.c:293]:
ERROR: receive_fd: EOF on 15
Mar 3 15:26:53 hostname kamailio[790]: ALERT: <core> [main.c:785]:
child process 796 exited by a signal 11
Mar 3 15:26:53 hostname kamailio[790]: ALERT: <core> [main.c:788]:
core was generated
Mar 3 15:26:53 hostname kamailio[790]: INFO: <core> [main.c:800]:
INFO: terminating due to SIGCHLD
Mar 3 15:26:53 hostname kamailio[802]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[799]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[801]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[798]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[800]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[797]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[792]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[803]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[795]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[791]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[794]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[793]: INFO: <core> [main.c:851]:
INFO: signal 15 received
Mar 3 15:26:53 hostname kamailio[790]: INFO: <core>
[mem/f_malloc.c:532]: freeing a free fragment (0x886b3ad8/0x886b3ae0)
- ignore
Mar 3 15:26:53 hostname kernel: [771135.681720] kamailio[790]:
segfault at 7dd193e0 ip 0816f5cd sp bf8acc20 error 4 in
kamailio[8048000+1ea000]
Currently we have fixed it using a custom route
route(T_REPLY);
.
.
.
route[T_REPLY] {
sl_send_reply("404","No route to destination");
}
This works correctly with no error or a warning message or a segfault.
We are using kamilio version: kamailio 3.3.4 (i386/linux) f8c8f2
Is this behavior a bug or should I dig deeper into the tm documentation?
Kind regards
Efelin
