[Users] t_on_failure doesn't
Bogdan-Andrei Iancu
bogdan at voice-system.ro
Tue Aug 22 15:29:56 CEST 2006
for the moment is not possible - please fill in a bug report on the
tracker to have in mind to work out a solution for this.
regards,
bogdan
Nathan Hawkins wrote:
>So am I correct in understanding that there is currently no way to
>handle this condition? This seems like a serious problem with using TCP.
>
>Bogdan-Andrei Iancu wrote:
>
>
>>Hi Nathan,
>>
>>failure route is called in case of a call failure (negative reply) and
>>not if a sending error occurred.
>>
>>the right way to do it will be to have a way to disable the auto reply
>>in case of error in t_reply() and deal with the error from the script.
>>
>>regards,
>>bogdan
>>
>>Nathan Hawkins wrote:
>>
>>
>>
>>>I'm trying to use the LCR module to do a failover between two
>>>gateways. For testing purposes, I have a gateway record pointed at an
>>>IP that isn't used. The problem is, I can't seem to get OpenSER to
>>>fail over to the other gateway. I want to use TCP (with TLS when I'm
>>>done), but I get this:
>>>
>>>Jul 20 15:36:57 asiago /usr/sbin/openser[11896]: route[6]: routing to
>>>INVITE sip:MYDID at MYIPADDRESS:5060;transport=tcp
>>>Jul 20 15:37:17 asiago /usr/sbin/openser[11896]: ERROR:
>>>tcp_blocking_connect: timeout (10)
>>>Jul 20 15:37:17 asiago /usr/sbin/openser[11896]: ERROR:
>>>tcpconn_connect: tcp_blocking_connect failed
>>>Jul 20 15:37:17 asiago /usr/sbin/openser[11896]: ERROR: tcp_send:
>>>connect failed
>>>Jul 20 15:37:17 asiago /usr/sbin/openser[11896]: msg_send: ERROR:
>>>tcp_send failed
>>>Jul 20 15:37:17 asiago /usr/sbin/openser[11896]:
>>>ERROR:tm:t_forward_nonack: sending request failed
>>>
>>>(IP and username censored.)
>>>
>>>OpenSER sends this to the caller:
>>>SIP/2.0 477 Unfortunately error on sending to next hop occurred (477/TM)
>>>
>>>Why can't I catch this with either the return code from t_relay or
>>>the failure route?
>>>
>>>It will go to the failure route if I get a reply back from the next
>>>hop. But I need to fail to the other gateway.
>>>
>>>The relevent parts of the configs are shown below. (Addresses censored.)
>>>
>>>Any help would greatly appreciated.
>>>
>>>Thanks,
>>>
>>> ---Nathan
>>>
>>>
>>>Config snippet:
>>>
>>> # ok, so maybe try a gateway...
>>> if( load_gws() ) {
>>> xlog("L_INFO","route[3]: gateway routing $rm $ru");
>>> t_on_failure("1");
>>> route(6);
>>> return;
>>> }
>>>}
>>>
>>>route[6] {
>>> xlog("L_INFO","route[6]: $rm $ru");
>>> t_on_failure("1");
>>> t_on_reply("1");
>>> if (!next_gw()) {
>>> xlog("L_INFO","route[6]: no more gateways for $ru");
>>> exit;
>>> }
>>> xlog("L_INFO","route[6]: routing to $rm $ru");
>>> if(!t_relay()) {
>>> xlog("L_INFO","route[6]: t_relay returned $rc");
>>> sl_reply_error();
>>> }
>>> return;
>>>}
>>>
>>>failure_route[1] {
>>> xlog("L_INFO","failure_route[1]: $rm $ru");
>>> t_on_failure("1");
>>> if(t_check_status("404")) {
>>> xlog("L_INFO","failure_route[1]: failing call with 404 for $ru");
>>> return;
>>> }
>>> if (!next_gw()) {
>>> xlog("L_INFO","route[6]: no more gateways for $ru");
>>> exit;
>>> }
>>> append_branch();
>>> xlog("L_INFO","failure_route[1]: sending $rm $ru");
>>> if(!t_relay()) {
>>> xlog("L_INFO","route[6]: t_relay returned $rc");
>>> }
>>> return;
>>>}
>>>
>>>
>>>Gateways:
>>>
>>>mysql> select * from gw where grp_id=4;
>>>+---------+--------+-----------+------+------------+-----------+--------+
>>>
>>>| gw_name | grp_id | ip_addr | port | uri_scheme | transport |
>>>prefix |
>>>+---------+--------+-----------+------+------------+-----------+--------+
>>>
>>>| ithaka | 4 | CENSORED | 5060 | NULL | 2 | NULL |
>>>| larink | 4 | BAD_ADDRESS | 5060 | NULL | 2 |
>>>NULL |
>>>+---------+--------+-----------+------+------------+-----------+--------+
>>>
>>>2 rows in set (0.00 sec)
>>>
>>>
>>>ithaka is a working OpenSER proxy. larink is an empty IP address.
>>>(There used to be a server named larink.)
>>>
>>>
>>>_______________________________________________
>>>Users mailing list
>>>Users at openser.org
>>>http://openser.org/cgi-bin/mailman/listinfo/users
>>>
>>>
>>>
>
>
>
More information about the Users
mailing list