[OpenSIPS-Users] ACC module with JSON events

[SamyGo]

Hi,

I'm using ACC module to trigger accounting events which land up in
event_xmlrpc and ultimately shipped to remote CDR collection server. The
sent data is obviously an XML - now, can I have an event_json like module
so data is sent as JSON or is there any other way I can achieve the same ?

Regards
Sammy
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] OpenSIPS crashing on 200 OK

[Ben Newlin]

Bogdan,

Here is the backtrace: https://pastebin.com/6tZhrWek

I am not using TCP, BIN, or HEP in this configuration. I am not using anything, 
to my knowledge, which should be causing TCP worker threads to be spawned.

Ben Newlin

From: Bogdan-Andrei Iancu 
Date: Thursday, February 22, 2018 at 5:26 AM
To: OpenSIPS users mailling list , Ben Newlin 

Subject: Re: [OpenSIPS-Users] OpenSIPS crashing on 200 OK

Hi Ben,

Do you have a full backtrace extracted from the core ?

In regards to the TCP processes, the TCP layer will automatically enabled 
itself depending on what protocols (tcp-based) you are using in script. Maybe 
you do not use SIP with TCP, but do you use BIN or HEP ?

Regards,


Bogdan-Andrei Iancu



OpenSIPS Founder and Developer

  http://www.opensips-solutions.com

OpenSIPS Summit 2018

  
http://www.opensips.org/events/Summit-2018Amsterdam
On 02/22/2018 12:38 AM, Ben Newlin wrote:
Hi,

I am currently upgrading to 2.3.3 and am running off the HEAD of the 2.3 
branch. My OpenSIPS is consistently crashing when trying to process the ACK to 
an initial INVITE request. I am testing with SIPp using a simple single call 
scenario. I see the following in the logs:

Feb 21 21:32:27 [383] DBG:core:io_wait_loop_epoll: [TCP_main] EPOLLHUP on IN 
->connection closed by the remote peer!
Feb 21 21:32:27 [383] CRITICAL:core:receive_fd: EOF on 33
Feb 21 21:32:27 [383] DBG:core:handle_worker: dead child 9, pid 370 (shutting 
down?)
Feb 21 21:32:27 [383] DBG:core:io_watch_del: [TCP_main] io_watch_del op on 
index 8 33 (0x8d4e80, 33, 8, 0x0,0x1) fd_no=29 called
Feb 21 21:32:27 [361] DBG:core:handle_sigs: status = 139
Feb 21 21:32:27 [361] INFO:core:handle_sigs: child process 370 exited by a 
signal 11
Feb 21 21:32:27 [361] INFO:core:handle_sigs: core was generated
Feb 21 21:32:27 [361] INFO:core:handle_sigs: terminating due to SIGCHLD

I also notice that OpenSIPS is starting up a number of TCP processes even 
though I do not use TCP in my script. I do not even have the proto_tcp module 
loaded. I used to use the “disable_tcp” global parameter in 1.11, but that has 
been removed in 2.3.

Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP receiver 
TCP"
Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP receiver 
TCP"
Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP receiver 
TCP"
Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP receiver 
TCP"
Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP receiver 
TCP"
Feb 21 22:03:56 [375] DBG:core:io_watch_add: [TCP_worker] io_watch_add op (3 on 
53) (0x8d4e80, 3, 1, (nil),1), fd_no=0/209715
Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP receiver 
TCP"
Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP receiver 
TCP"
Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP receiver 
TCP"

Initialization logs: 
https://pastebin.com/nsJQGkdy
Call logs: https://pastebin.com/vY5crK8U

I also have the core dump if any more information is needed from that.

Thanks,
Ben Newlin





___

Users mailing list

Users@lists.opensips.org

http://lists.opensips.org/cgi-bin/mailman/listinfo/users


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] del_uri_param failure

[Ben Newlin]

Bogdan,

Thanks for the quick response. I have verified the fix works in 2.3 branch head.

Thanks,
Ben Newlin


From: Bogdan-Andrei Iancu 
Date: Thursday, February 22, 2018 at 10:03 AM
To: Ben Newlin , OpenSIPS users mailling list 

Subject: Re: [OpenSIPS-Users] del_uri_param failure

Hi Ben,

I just pushed the fix on git, it was a really stupid error.

Thanks for reporting,


Bogdan-Andrei Iancu



OpenSIPS Founder and Developer

  http://www.opensips-solutions.com

OpenSIPS Summit 2018

  
http://www.opensips.org/events/Summit-2018Amsterdam
On 02/22/2018 03:18 PM, Ben Newlin wrote:
I saw this behavior in 2.3.2, 2.3.3, and I am now running on HEAD of the 2.3 
branch.

Ben Newlin


From: Bogdan-Andrei Iancu 
Date: Thursday, February 22, 2018 at 5:20 AM
To: OpenSIPS users mailling list 
, Ben Newlin 

Subject: Re: [OpenSIPS-Users] del_uri_param failure

Hi Ben,

What OpensSIPS version and revision are you working with ?

Regards,



Bogdan-Andrei Iancu



OpenSIPS Founder and Developer

  http://www.opensips-solutions.com

OpenSIPS Summit 2018

  
http://www.opensips.org/events/Summit-2018Amsterdam
On 02/22/2018 01:07 AM, Ben Newlin wrote:
Hi,

I am very glad to have the new del_uri_param function. This was a common 
problem of mine and it is great not to have to use regex to do this. However, 
while implementing this I have run into some strange behavior by the function 
when the URI param being deleted does not exist.

In my case I am using the dialog module and attempting to remove a URI param 
just after the dialog creation. When the function does not find the URI param, 
it causes the dialog to immediately be destroyed and all message processing 
stops, including exiting the script.

Feb 21 22:47:42 [371] DBG:uri:del_uri_param: requested key not found in RURI
Feb 21 22:47:42 [371] DBG:dialog:next_state_dlg: unref dlg 0x7f460780a8c8 with 
1 -> 2 in entry 0x7f46077fbfa8
Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding string param
Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding string param
Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding int param
Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding int param
Feb 21 22:47:42 [371] DBG:core:destroy_avp_list: destroying list (nil)
Feb 21 22:47:42 [371] DBG:dialog:next_state_dlg: dialog 0x7f460780a8c8 changed 
from state 1 to state 5, due event 1
Feb 21 22:47:42 [371] DBG:dialog:dlg_onreply: dialog 0x7f460780a8c8 failed 
(negative reply)
Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: unref dlg 0x7f460780a8c8 with 1 -> 
1 in entry 0x7f46077fbfa8
Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: unref dlg 0x7f460780a8c8 with 1 -> 
0 in entry 0x7f46077fbfa8
Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: ref <=0 for dialog 0x7f460780a8c8
Feb 21 22:47:42 [371] DBG:dialog:destroy_dlg: destroying dialog 0x7f460780a8c8
Feb 21 22:47:42 [371] DBG:dialog:destroy_dlg: dlg expired or not in list - dlg 
0x7f460780a8c8 [3710:1818203549] with clid 
'2-185@127.0.0.1' and tags '185SIPpTag002' 'NULL'
Feb 21 22:47:42 [371] DBG:core:destroy_avp_list: destroying list 0x7f460780c048
Feb 21 22:47:42 [371] DBG:core:receive_msg: cleaning up

The logs indicate that a DLG_EVENT_TDEL is being raised which, when the dialog 
is still in UNCONFIRMED state, causes the dialog to be destroyed. It’s not 
clear to me how or why the del_uri_param function could be doing this, 
especially as a transaction hasn’t even been created for the message yet in 
this case. I’m not sure what effect this would have if the dialog is in other 
states or at other times during the call.

It took me a while to realize it was the del_uri_param function causing this, 
as it seems so strange. But I have verified that when I remove the function, or 
when I verify the URI param exists before calling the function, everything is 
fine. That workaround works perfectly well, but it seemed such strange and 
catastrophic error behavior to drop the entire message that I wanted to report 
it anyway to see if anything needed to be addressed.

Call traces can be found here: 
https://pastebin.com/9FnmJCD9

You will see the same INVITE is offered multiple times as OpenSIPS is not 
responding after dropping the previous requests and dialog.

Thanks,
Ben Newlin






___

Users mailing list

Users@lists.opensips.org

http://lists.opensips.org/cgi-bin/mailman/listinfo/users





___
Users mailing 

Re: [OpenSIPS-Users] Problem with simultaneous CANCEL + 200 OK

[Bogdan-Andrei Iancu]

Daniel,

The pcap shows only one leg of the the communication (I guess caller 
versus OpenSIPS). So I cannot see what happens on the callee side. 
Still, I see the first cancel is rejected with 400 (I guess by OpenSIPS 
- check the logs to see the reason) and the second one is accepted by 
OpenSIPS (no idea if relayed to callee or not). Still, if there is a 
race between the cancel from caller and the answer from callee, it is up 
to the caller to sort it outaccording to RFC, the caller must ACK 
the received 200 OK (even if CANCEL was sent) and if it really wants to 
terminate the call, it has to fire a BYE.


Regards,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
  http://www.opensips-solutions.com
OpenSIPS Summit 2018
  http://www.opensips.org/events/Summit-2018Amsterdam

On 02/22/2018 04:22 PM, Daniel Zanutti wrote:

Hi Bogdan

Thanks for replying.

Here is the PCAP, please take a look: 
https://drive.google.com/file/d/1e7SKjxDtdVYmN-7fCHSEqNEsNHjPsaKo/view?usp=sharing


Thanks

On Thu, Feb 22, 2018 at 7:30 AM, Bogdan-Andrei Iancu 
> wrote:


Hi Daniel,

Without a pcap showing the signaling is hard to understand (not to
mention helping) your scenario. Please provide a link to the pcap
or ngrep.

Regards,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
   http://www.opensips-solutions.com 
OpenSIPS Summit 2018
   http://www.opensips.org/events/Summit-2018Amsterdam


On 02/20/2018 06:43 PM, Daniel Zanutti wrote:

Hey

I had a problem when receiving simultaneous CANCEL from customer
and 200 OK from gateway.

Seems that the first CANCEL was rejected, but the second CANCEL
was accepted. This second CANCEL did NOT go to the gateway, just
Opensips received and replied with 200 OK.

This is the log of the first CANCEL:
Feb 15 18:39:22 /sbin/opensips[28845]: SCRIPT:TRAFFIC:WARNING:
method CANCEL (7Qbq3O3CReMfPflAtl8NY3ddTqPVBHO2785126@2.2.2.2/
XAeG2xj278512T2
/ 1839212581509953 ) not validated and not fixed ( code=-1 )

code -1 is the return of validate_dialog()

Second CANCEL didn't generated a log.

Shouldn't all CANCELs be rejected? On this case, just the first
one was rejected.

I'm using version 1.9.11.




___
Users mailing list
Users@lists.opensips.org 
http://lists.opensips.org/cgi-bin/mailman/listinfo/users






___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] del_uri_param failure

[Bogdan-Andrei Iancu]

Hi Ben,

I just pushed the fix on git, it was areally stupid error.

Thanks for reporting,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
  http://www.opensips-solutions.com
OpenSIPS Summit 2018
  http://www.opensips.org/events/Summit-2018Amsterdam

On 02/22/2018 03:18 PM, Ben Newlin wrote:


I saw this behavior in 2.3.2, 2.3.3, and I am now running on HEAD of 
the 2.3 branch.


Ben Newlin

*From: *Bogdan-Andrei Iancu 
*Date: *Thursday, February 22, 2018 at 5:20 AM
*To: *OpenSIPS users mailling list , Ben 
Newlin 

*Subject: *Re: [OpenSIPS-Users] del_uri_param failure

Hi Ben,

What OpensSIPS version and revision are you working with ?

Regards,

Bogdan-Andrei Iancu
OpenSIPS Founder and Developer
http://www.opensips-solutions.com
OpenSIPS Summit 2018
http://www.opensips.org/events/Summit-2018Amsterdam

On 02/22/2018 01:07 AM, Ben Newlin wrote:

Hi,

I am very glad to have the new del_uri_param function. This was a
common problem of mine and it is great not to have to use regex to
do this. However, while implementing this I have run into some
strange behavior by the function when the URI param being deleted
does not exist.

In my case I am using the dialog module and attempting to remove a
URI param just after the dialog creation. When the function does
not find the URI param, it causes the dialog to immediately be
destroyed and all message processing stops, including exiting the
script.

Feb 21 22:47:42 [371] DBG:uri:del_uri_param: requested key not
found in RURI

Feb 21 22:47:42 [371] DBG:dialog:next_state_dlg: unref dlg
0x7f460780a8c8 with 1 -> 2 in entry 0x7f46077fbfa8

Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding string param

Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding string param

Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding int param

Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding int param

Feb 21 22:47:42 [371] DBG:core:destroy_avp_list: destroying list (nil)

Feb 21 22:47:42 [371] DBG:dialog:next_state_dlg: dialog
0x7f460780a8c8 changed from state 1 to state 5, due event 1

Feb 21 22:47:42 [371] DBG:dialog:dlg_onreply: dialog
0x7f460780a8c8 failed (negative reply)

Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: unref dlg
0x7f460780a8c8 with 1 -> 1 in entry 0x7f46077fbfa8

Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: unref dlg
0x7f460780a8c8 with 1 -> 0 in entry 0x7f46077fbfa8

Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: ref <=0 for dialog
0x7f460780a8c8

Feb 21 22:47:42 [371] DBG:dialog:destroy_dlg: destroying dialog
0x7f460780a8c8

Feb 21 22:47:42 [371] DBG:dialog:destroy_dlg: dlg expired or not
in list - dlg 0x7f460780a8c8 [3710:1818203549] with clid
'2-185@127.0.0.1' and tags '185SIPpTag002'
'NULL'

Feb 21 22:47:42 [371] DBG:core:destroy_avp_list: destroying list
0x7f460780c048

Feb 21 22:47:42 [371] DBG:core:receive_msg: cleaning up

The logs indicate that a DLG_EVENT_TDEL is being raised which,
when the dialog is still in UNCONFIRMED state, causes the dialog
to be destroyed. It’s not clear to me how or why the del_uri_param
function could be doing this, especially as a transaction hasn’t
even been created for the message yet in this case. I’m not sure
what effect this would have if the dialog is in other states or at
other times during the call.

It took me a while to realize it was the del_uri_param function
causing this, as it seems so strange. But I have verified that
when I remove the function, or when I verify the URI param exists
before calling the function, everything is fine. That workaround
works perfectly well, but it seemed such strange and catastrophic
error behavior to drop the entire message that I wanted to report
it anyway to see if anything needed to be addressed.

Call traces can be found here: https://pastebin.com/9FnmJCD9

You will see the same INVITE is offered multiple times as OpenSIPS
is not responding after dropping the previous requests and dialog.

Thanks,

Ben Newlin




___

Users mailing list

Users@lists.opensips.org

http://lists.opensips.org/cgi-bin/mailman/listinfo/users





___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Problem with simultaneous CANCEL + 200 OK

[Daniel Zanutti]

Hi Bogdan

Thanks for replying.

Here is the PCAP, please take a look:
https://drive.google.com/file/d/1e7SKjxDtdVYmN-7fCHSEqNEsNHjPsaKo/view?usp=sharing

Thanks

On Thu, Feb 22, 2018 at 7:30 AM, Bogdan-Andrei Iancu 
wrote:

> Hi Daniel,
>
> Without a pcap showing the signaling is hard to understand (not to mention
> helping) your scenario. Please provide a link to the pcap or ngrep.
>
> Regards,
>
> Bogdan-Andrei Iancu
>
> OpenSIPS Founder and Developer
>   http://www.opensips-solutions.com
> OpenSIPS Summit 2018
>   http://www.opensips.org/events/Summit-2018Amsterdam
>
> On 02/20/2018 06:43 PM, Daniel Zanutti wrote:
>
> Hey
>
> I had a problem when receiving simultaneous CANCEL from customer and 200
> OK from gateway.
>
> Seems that the first CANCEL was rejected, but the second CANCEL was
> accepted. This second CANCEL did NOT go to the gateway, just Opensips
> received and replied with 200 OK.
>
> This is the log of the first CANCEL:
> Feb 15 18:39:22 /sbin/opensips[28845]: SCRIPT:TRAFFIC:WARNING: method
> CANCEL ( 7Qbq3O3CReMfPflAtl8NY3ddTqPVBHO2785126@2.2.2.2/ XAeG2xj278512T2
> / 1839212581509953 ) not validated and not fixed ( code=-1 )
>
> code -1 is the return of validate_dialog()
>
> Second CANCEL didn't generated a log.
>
> Shouldn't all CANCELs be rejected? On this case, just the first one was
> rejected.
>
> I'm using version 1.9.11.
>
>
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] del_uri_param failure

[Ben Newlin]

I saw this behavior in 2.3.2, 2.3.3, and I am now running on HEAD of the 2.3 
branch.

Ben Newlin


From: Bogdan-Andrei Iancu 
Date: Thursday, February 22, 2018 at 5:20 AM
To: OpenSIPS users mailling list , Ben Newlin 

Subject: Re: [OpenSIPS-Users] del_uri_param failure

Hi Ben,

What OpensSIPS version and revision are you working with ?

Regards,


Bogdan-Andrei Iancu



OpenSIPS Founder and Developer

  http://www.opensips-solutions.com

OpenSIPS Summit 2018

  
http://www.opensips.org/events/Summit-2018Amsterdam
On 02/22/2018 01:07 AM, Ben Newlin wrote:
Hi,

I am very glad to have the new del_uri_param function. This was a common 
problem of mine and it is great not to have to use regex to do this. However, 
while implementing this I have run into some strange behavior by the function 
when the URI param being deleted does not exist.

In my case I am using the dialog module and attempting to remove a URI param 
just after the dialog creation. When the function does not find the URI param, 
it causes the dialog to immediately be destroyed and all message processing 
stops, including exiting the script.

Feb 21 22:47:42 [371] DBG:uri:del_uri_param: requested key not found in RURI
Feb 21 22:47:42 [371] DBG:dialog:next_state_dlg: unref dlg 0x7f460780a8c8 with 
1 -> 2 in entry 0x7f46077fbfa8
Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding string param
Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding string param
Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding int param
Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding int param
Feb 21 22:47:42 [371] DBG:core:destroy_avp_list: destroying list (nil)
Feb 21 22:47:42 [371] DBG:dialog:next_state_dlg: dialog 0x7f460780a8c8 changed 
from state 1 to state 5, due event 1
Feb 21 22:47:42 [371] DBG:dialog:dlg_onreply: dialog 0x7f460780a8c8 failed 
(negative reply)
Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: unref dlg 0x7f460780a8c8 with 1 -> 
1 in entry 0x7f46077fbfa8
Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: unref dlg 0x7f460780a8c8 with 1 -> 
0 in entry 0x7f46077fbfa8
Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: ref <=0 for dialog 0x7f460780a8c8
Feb 21 22:47:42 [371] DBG:dialog:destroy_dlg: destroying dialog 0x7f460780a8c8
Feb 21 22:47:42 [371] DBG:dialog:destroy_dlg: dlg expired or not in list - dlg 
0x7f460780a8c8 [3710:1818203549] with clid 
'2-185@127.0.0.1' and tags '185SIPpTag002' 'NULL'
Feb 21 22:47:42 [371] DBG:core:destroy_avp_list: destroying list 0x7f460780c048
Feb 21 22:47:42 [371] DBG:core:receive_msg: cleaning up

The logs indicate that a DLG_EVENT_TDEL is being raised which, when the dialog 
is still in UNCONFIRMED state, causes the dialog to be destroyed. It’s not 
clear to me how or why the del_uri_param function could be doing this, 
especially as a transaction hasn’t even been created for the message yet in 
this case. I’m not sure what effect this would have if the dialog is in other 
states or at other times during the call.

It took me a while to realize it was the del_uri_param function causing this, 
as it seems so strange. But I have verified that when I remove the function, or 
when I verify the URI param exists before calling the function, everything is 
fine. That workaround works perfectly well, but it seemed such strange and 
catastrophic error behavior to drop the entire message that I wanted to report 
it anyway to see if anything needed to be addressed.

Call traces can be found here: 
https://pastebin.com/9FnmJCD9

You will see the same INVITE is offered multiple times as OpenSIPS is not 
responding after dropping the previous requests and dialog.

Thanks,
Ben Newlin





___

Users mailing list

Users@lists.opensips.org

http://lists.opensips.org/cgi-bin/mailman/listinfo/users


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Problem with simultaneous CANCEL + 200 OK

[Bogdan-Andrei Iancu]

Hi Daniel,

Without a pcap showing the signaling is hard to understand (not to 
mention helping) your scenario. Please provide a link to the pcap or ngrep.


Regards,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
  http://www.opensips-solutions.com
OpenSIPS Summit 2018
  http://www.opensips.org/events/Summit-2018Amsterdam

On 02/20/2018 06:43 PM, Daniel Zanutti wrote:

Hey

I had a problem when receiving simultaneous CANCEL from customer and 
200 OK from gateway.


Seems that the first CANCEL was rejected, but the second CANCEL was 
accepted. This second CANCEL did NOT go to the gateway, just Opensips 
received and replied with 200 OK.


This is the log of the first CANCEL:
Feb 15 18:39:22 /sbin/opensips[28845]: SCRIPT:TRAFFIC:WARNING: method 
CANCEL (7Qbq3O3CReMfPflAtl8NY3ddTqPVBHO2785126@2.2.2.2/ 
XAeG2xj278512T2 
/ 1839212581509953 ) not validated and not fixed ( code=-1 )


code -1 is the return of validate_dialog()

Second CANCEL didn't generated a log.

Shouldn't all CANCELs be rejected? On this case, just the first one 
was rejected.


I'm using version 1.9.11.




___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] How to avoid/solve TCP blocked connection?

[Bogdan-Andrei Iancu]

Hi Rodrigo,

If I understand correctly, you get these errors (due failed connect 
attempt) when trying to deliver (from OpenSIPS) a call to a registered 
users who lost his TCP connection, right ?


Regards,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
  http://www.opensips-solutions.com
OpenSIPS Summit 2018
  http://www.opensips.org/events/Summit-2018Amsterdam

On 02/20/2018 05:07 PM, Rodrigo Pimenta Carvalho wrote:


Hi.


My softphone is registered with the following AOR:


-

AOR:: g1r2u3p4o5
Contact:: sip:g1r2u3p4o5@127.0.0.1:50353;transport=TLS;ob Q=
Expires:: 10
Callid:: 53e387dc-81fe-45f9-a6f1-8a5cf4248d62
Cseq:: 37190
User-agent:: n/a
Received:: sip:127.0.0.1:49678;transport=TLS
State:: CS_SYNC
Flags:: 0
Cflags:: NAT
Socket:: tls:127.0.0.1:5061
Methods:: 8063
Attr:: in_same_network
SIP_instance:: 
- 




Sometimes, when it receives a call and answers, the Opensips show the 
following error:



[21532] INFO:core:probe_max_sock_buff: using snd buffer of 320 kb
[21532] INFO:core:init_sock_keepalive: TCP keepalive enabled on socket 22
[21532] ERROR:core:tcp_connect_blocking: poll error: flags 28 - 4 8 16 32
[21532] ERROR:core:tcp_connect_blocking: failed to retrieve SO_ERROR 
[server=127.0.0.1:50353] (111) Connection refused

[21532] ERROR:proto_tls:tls_sync_connect: tcp_blocking_connect failed
[21532] ERROR:proto_tls:proto_tls_send: connect failed
[21532] ERROR:core:msg_send: send() for proto 3 failed

Is there a way to avoid this kind of problem? That is, can I configure 
the OpenSIPS to renew some TCP connection?


Any hint will be very helpful!

Best regards.


RODRIGO PIMENTA CARVALHO
Inatel Competence Center
Software
Ph: +55 35 3471 9200 RAMAL 979


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] OpenSIPS crashing on 200 OK

[Bogdan-Andrei Iancu]

Hi Ben,

Do you have a full backtrace extracted from the core ?

In regards to the TCP processes, the TCP layer will automatically 
enabled itself depending on what protocols (tcp-based) you are using in 
script. Maybe you do not use SIP with TCP, but do you use BIN or HEP ?


Regards,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
  http://www.opensips-solutions.com
OpenSIPS Summit 2018
  http://www.opensips.org/events/Summit-2018Amsterdam

On 02/22/2018 12:38 AM, Ben Newlin wrote:


Hi,

I am currently upgrading to 2.3.3 and am running off the HEAD of the 
2.3 branch. My OpenSIPS is consistently crashing when trying to 
process the ACK to an initial INVITE request. I am testing with SIPp 
using a simple single call scenario. I see the following in the logs:


Feb 21 21:32:27 [383] DBG:core:io_wait_loop_epoll: [TCP_main] EPOLLHUP 
on IN ->connection closed by the remote peer!


Feb 21 21:32:27 [383] CRITICAL:core:receive_fd: EOF on 33

Feb 21 21:32:27 [383] DBG:core:handle_worker: dead child 9, pid 370 
(shutting down?)


Feb 21 21:32:27 [383] DBG:core:io_watch_del: [TCP_main] io_watch_del 
op on index 8 33 (0x8d4e80, 33, 8, 0x0,0x1) fd_no=29 called


Feb 21 21:32:27 [361] DBG:core:handle_sigs: status = 139

Feb 21 21:32:27 [361] INFO:core:handle_sigs: child process 370 exited 
by a signal 11


Feb 21 21:32:27 [361] INFO:core:handle_sigs: core was generated

Feb 21 21:32:27 [361] INFO:core:handle_sigs: terminating due to SIGCHLD

I also notice that OpenSIPS is starting up a number of TCP processes 
even though I do not use TCP in my script. I do not even have the 
proto_tcp module loaded. I used to use the “disable_tcp” global 
parameter in 1.11, but that has been removed in 2.3.


Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP 
receiver TCP"


Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP 
receiver TCP"


Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP 
receiver TCP"


Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP 
receiver TCP"


Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP 
receiver TCP"


Feb 21 22:03:56 [375] DBG:core:io_watch_add: [TCP_worker] io_watch_add 
op (3 on 53) (0x8d4e80, 3, 1, (nil),1), fd_no=0/209715


Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP 
receiver TCP"


Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP 
receiver TCP"


Feb 21 22:03:56 [361] DBG:core:internal_fork: forking new process "SIP 
receiver TCP"


Initialization logs: https://pastebin.com/nsJQGkdy

Call logs: https://pastebin.com/vY5crK8U

I also have the core dump if any more information is needed from that.

Thanks,

Ben Newlin



___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] del_uri_param failure

[Bogdan-Andrei Iancu]

Hi Ben,

What OpensSIPS version and revision are you working with ?

Regards,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
  http://www.opensips-solutions.com
OpenSIPS Summit 2018
  http://www.opensips.org/events/Summit-2018Amsterdam

On 02/22/2018 01:07 AM, Ben Newlin wrote:


Hi,

I am very glad to have the new del_uri_param function. This was a 
common problem of mine and it is great not to have to use regex to do 
this. However, while implementing this I have run into some strange 
behavior by the function when the URI param being deleted does not exist.


In my case I am using the dialog module and attempting to remove a URI 
param just after the dialog creation. When the function does not find 
the URI param, it causes the dialog to immediately be destroyed and 
all message processing stops, including exiting the script.


Feb 21 22:47:42 [371] DBG:uri:del_uri_param: requested key not found 
in RURI


Feb 21 22:47:42 [371] DBG:dialog:next_state_dlg: unref dlg 
0x7f460780a8c8 with 1 -> 2 in entry 0x7f46077fbfa8


Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding string param

Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding string param

Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding int param

Feb 21 22:47:42 [371] DBG:core:evi_param_set: adding int param

Feb 21 22:47:42 [371] DBG:core:destroy_avp_list: destroying list (nil)

Feb 21 22:47:42 [371] DBG:dialog:next_state_dlg: dialog 0x7f460780a8c8 
changed from state 1 to state 5, due event 1


Feb 21 22:47:42 [371] DBG:dialog:dlg_onreply: dialog 0x7f460780a8c8 
failed (negative reply)


Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: unref dlg 0x7f460780a8c8 
with 1 -> 1 in entry 0x7f46077fbfa8


Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: unref dlg 0x7f460780a8c8 
with 1 -> 0 in entry 0x7f46077fbfa8


Feb 21 22:47:42 [371] DBG:dialog:unref_dlg: ref <=0 for dialog 
0x7f460780a8c8


Feb 21 22:47:42 [371] DBG:dialog:destroy_dlg: destroying dialog 
0x7f460780a8c8


Feb 21 22:47:42 [371] DBG:dialog:destroy_dlg: dlg expired or not in 
list - dlg 0x7f460780a8c8 [3710:1818203549] with clid 
'2-185@127.0.0.1' and tags '185SIPpTag002' 'NULL'


Feb 21 22:47:42 [371] DBG:core:destroy_avp_list: destroying list 
0x7f460780c048


Feb 21 22:47:42 [371] DBG:core:receive_msg: cleaning up

The logs indicate that a DLG_EVENT_TDEL is being raised which, when 
the dialog is still in UNCONFIRMED state, causes the dialog to be 
destroyed. It’s not clear to me how or why the del_uri_param function 
could be doing this, especially as a transaction hasn’t even been 
created for the message yet in this case. I’m not sure what effect 
this would have if the dialog is in other states or at other times 
during the call.


It took me a while to realize it was the del_uri_param function 
causing this, as it seems so strange. But I have verified that when I 
remove the function, or when I verify the URI param exists before 
calling the function, everything is fine. That workaround works 
perfectly well, but it seemed such strange and catastrophic error 
behavior to drop the entire message that I wanted to report it anyway 
to see if anything needed to be addressed.


Call traces can be found here: https://pastebin.com/9FnmJCD9

You will see the same INVITE is offered multiple times as OpenSIPS is 
not responding after dropping the previous requests and dialog.


Thanks,

Ben Newlin



___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Avpops and db_virtual modules

[Bogdan-Andrei Iancu]

Cool, Ihappy this issue was fixed. Waiting for your new report :)

Regards,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
  http://www.opensips-solutions.com
OpenSIPS Summit 2018
  http://www.opensips.org/events/Summit-2018Amsterdam

On 02/21/2018 09:57 PM, Ben Newlin wrote:


Bogdan,

I have tested with this fix and I was able to start up with these 
modules properly. I am getting another error now, but I will report 
that in a new thread so as not to confuse unrelated issues.


Thanks,

Ben Newlin

*From: *Bogdan-Andrei Iancu 
*Date: *Wednesday, February 21, 2018 at 9:53 AM
*To: *Ben Newlin , OpenSIPS users mailling 
list 

*Subject: *Re: [OpenSIPS-Users] Avpops and db_virtual modules

Hi Ben,

OK, I reviewed the changes and the code and I identified the problem 
in the way db_virtual is computing the capabilities in ROUND ROBIN 
mode - more or less it was discarding the UPDATE DB capabilities, 
which was required by avpops.


I pushed a fix for this, this is for 2.3:
https://github.com/OpenSIPS/opensips/commit/edded2f1431cfaaf75ed2af0c5ff9ec28548e7c5

Please update and give it a try !

Thanks and Regards,

Bogdan-Andrei Iancu
OpenSIPS Founder and Developer
   http://www.opensips-solutions.com
OpenSIPS Summit 2018
   http://www.opensips.org/events/Summit-2018Amsterdam

On 02/19/2018 04:20 PM, Ben Newlin wrote:

Bogdan,

Thank you for looking into this. I have confirmed that it works in
2.3.2, which was released in 8/17 before the commit you provided.

So is there really an issue with the db_virtual capabilities that
was masked by this bug, or is db_virtual simply not publishing the
correct capabilities itself?

Thanks,

Ben Newlin

*From: *Bogdan-Andrei Iancu 

*Date: *Monday, February 19, 2018 at 6:11 AM
*To: *Ben Newlin 
, OpenSIPS users mailling list
 
*Subject: *Re: [OpenSIPS-Users] Avpops and db_virtual modules

Hi Ben,

Could you check if a 2.3 version prior this commit:

https://github.com/OpenSIPS/opensips/commit/f25fbf631b6fb6eafea192f95f9a43566d68cc46#diff-91bbef43b346f11cbef768a1b01c74e0
will still have the 1.11 behavior ?







___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] [Livestream][ClueCon Weekly] OpenSIPS-FreeSWITCH integration discussion - Feb 21st @ 12:00 PM CST

[Liviu Chircu]

Thank you to everyone who took part in the livestream yesterday! You may 
find its recording here [1].


As a final note, we would once again like to give proper credits to Mr. 
Giovanni Maruzzelli, who provided both the initial idea and some 
proof-of-concept code samples for the extended OpenSIPS-FreeSWITCH 
integration.


Best regards,

[1]: https://www.youtube.com/watch?v=9a2Cd1AV7V0

Liviu Chircu
OpenSIPS Developer
http://www.opensips-solutions.com

On 20.02.2018 21:22, Liviu Chircu wrote:

Hi all,

OpenSIPS is going to be live on the FreeSWITCH "ClueCon Weekly" 
conference tomorrow, February 21st @ 12 PM CST!


On this edition, the duty of representing the project falls on my 
shoulders. The main topic of discussion will revolve around the latest 
FreeSWITCH integration features that we've pushed into the development 
branch. So if you also use FreeSWITCH but you haven't caught up with 
them yet [1], tomorrow would be a good time to tune in and hear all 
about it. Towards the end, there will also be a Q/A section.


See you there!

Best regards,

[1]: 
https://blog.opensips.org/2018/01/17/how-to-script-advanced-freeswitch-integrations-with-opensips-2-4/





___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users