- Nov 16, 2018
-
-
George Joseph authored
-
- Nov 15, 2018
-
-
George Joseph authored
-
Corey Farrell authored
We cannot use need_el_end and SIGURG when restarting. Instead we need to run el_end within the SIGHUP restartnow handler. ASTERISK-28158 Change-Id: Ia852276363c81bdcf1aa29eb4558c5c2fa1218a0
-
- Nov 14, 2018
-
-
George Joseph authored
When dn_expand was being called on SRV and NAPTR results, the return value was being used to calculate the size of the buffer needed to store the host names. Since dn_expand returns the length of the COMPRESSED name the buffer could be too short to hold the EXPANDED name. The expanded name is NULL terminated so using strlen() is the correct way to determine the length actually needed for the buffer. ASTERISK-28127 Reported by: Jan Hoffmann patches: patch.diff submitted by janhoffmann (license 6986) Change-Id: I4d35d6c431c6c6836cb61d37b1378cc47f0b414d
-
Joshua Colp authored
-
Joshua Colp authored
-
- Nov 13, 2018
-
-
George Joseph authored
-
- Nov 12, 2018
-
-
Corey Farrell authored
Merge storage for the stats object and name string into the main allocation for struct ast_taskprocessor. Change-Id: I74fe9a7f357f0e6d63152f163cf5eef6428218e1
-
Joshua Colp authored
-
- Nov 11, 2018
-
-
Corey Farrell authored
* Ignore console=yes configuration option in remote console processes. * Use new flag to tell consolethread to run el_end and exit when needed. ASTERISK-28158 Change-Id: I9e23b31d4211417ddc88c6bbfd83ea4c9f3e5438
-
- Nov 08, 2018
-
-
Corey Farrell authored
pack_string crashed on non-NULL strings returned when s->has_error was true if the string was the result of 's' format without '#', '%' or '+'. Change-Id: Ic125df691d81ba2cbc413e37bdae657b304d20d0
-
Corey Farrell authored
If multiple [globals] sections are used (for example via separate included files), only the first one is processed. This can result in lost global variables when using a modular extensions.conf. ASTERISK-28146 #close Change-Id: Iaac810c0a7c4d9b1bf8989fcc041cdb910ef08a0
-
- Nov 07, 2018
-
-
Chris-Savinovich authored
When Asterisk's taskprocessors get overloaded we need to reduce the work load. res_pjsip currently ignores new SIP requests and relies on SIP retransmissions in the hope that the overload condition will clear soon enough to handle the retransmitted SIP request. This change adds the following code after ast_taskprocessor_alert_get() has returned TRUE: 1- identifies transport type. If non-udp then send a 503 response 2- if transport type is udp/udp6 then ignore, as before. Change-Id: I1c230b40d43a254ea0f226b7acf9ee480a5d3836
-
Joshua Colp authored
-
- Nov 06, 2018
-
-
Kevin Harwell authored
The use of a '|' in the "global/debug" synopsis documentation caused the generated html table on the wiki to add an extra column that included the text after the pipe. This patch replaces the pipe with a comma. ASTERISK-28150 Change-Id: I3d79a6ca6d733d9cb290e779438114884b98a719
-
Alexei Gradinari authored
The current round-robin method does not take the current taskprocessor load into consideration when distributing requests. Using the least-size method the request goes to the taskprocessor that is servicing the least number of active tasks at the current time. Longer running tasks with the round-robin method can delay processing tasks. * Change the algorithm from round-robin to least-size for picking the PJSIP taskprocessor from the default serializer pool. Change-Id: I7b8d8cc2c2490494f579374b6af0a4868e3a37cd
-
- Nov 05, 2018
-
-
George Joseph authored
-
Joshua Colp authored
As mentioned in the comment I've added in the code there is no ability to unsubscribe all subscribers from a topic and explicitly destroy it. This is not currently a problem as we have two types of topics: Long lived topics which exist for the lifetime of the system. Ephemeral topics which feed a long lived topic. In the case of the ephemeral topics there is no subscriber which does not have its lifetime managed by the same entity that has created the topic. This ensures that when the topic is being unreferenced the subscribers are also unsubscribed and destroyed, allowing the topic to ultimately be destroyed as well. Change-Id: Ic5e244da7b16b1895ba1fc5ece481ebba5809c9a
-
- Nov 02, 2018
-
-
Jasper Hafkenscheid authored
When a call pickup is performed using and invite with replaces header the ast_do_pickup method is attempted and a PICKUP stasis message is sent. ASTERISK-28081 #close Reported-by: Luit van Drongelen Change-Id: Ieb1442027a3ce6ae55faca47bc095e53972f947a
-
- Nov 01, 2018
-
-
Pascal Cadotte Michaud authored
Using the --quiet or -q option in conjonction with /dev/stdout as the output file allow the output to be used as a valid configuration. Given a script that generates a valid sip.conf I can pipe the output of that script into `sip_to_pjsip.py -q /dev/stdin /dev/stdout`. This allow me to use that piped command in my pjsip.conf using the `exec` command. ASTERISK-28136 Change-Id: I7b0e2e90e2549f3f8e01dc96701f111b5874c88d
-
- Oct 31, 2018
-
-
George Joseph authored
-
George Joseph authored
-
George Joseph authored
-
George Joseph authored
-
George Joseph authored
-
Joshua Colp authored
ASTERISK-28087 Change-Id: I69d48813ec514f5ef06c6de994cba52630e0a3b4
-
- Oct 30, 2018
-
-
Richard Mudgett authored
ASTERISK-28087 Change-Id: I046d018015427d0916fab571b5a4f5367476f729
-
Alexei Gradinari authored
This patch adds new options 'trust_connected_line' and 'send_connected_line' to the endpoint. The option 'trust_connected_line' is to control if connected line updates are accepted from this endpoint. The option 'send_connected_line' is to control if connected line updates can be sent to this endpoint. The default value is 'yes' for both options. Change-Id: I16af967815efd904597ec2f033337e4333d097cd
-
Pascal Cadotte Michaud authored
Given a sip.conf with the following content: setvar FOO=1 setvar BAR=42 I want my generated pjsip.conf to containt the following set_vars set_var FOO=1 set_var BAR=42 in the matching endpoint section. Change-Id: I6c822401fda4133c3b44bf31e655b4eb939d4d26
-
- Oct 29, 2018
-
-
George Joseph authored
-
- Oct 27, 2018
-
-
Alexei Gradinari authored
The module 'res_pjsip_notify' inefficiently makes a lot of DB requests on CLI completion on the endpoint. For example if there are 10k endpoints the module makes 10k requests of these 10k records. Even if a part of the endpoint entered the module makes the same 10k requests and then filtered them by itself. This patch gathers endpoints container by prefix and adds all gathered endpoints to completion at once. ASTERISK-28137 #close Change-Id: Ic20024912cc77bf4d3e476c4cd853293c52b254b
-
- Oct 26, 2018
-
-
Torrey Searle authored
Add a new global flag to res_pjsip to allow the callerid to be used as the username in the contact header. This allows chan_pjsip to have the same behavour as chan_sip ASTERISK-28087 #close Change-Id: I9a720e058323f6862a91c62f8a8c1a4b5c087b95
-
- Oct 25, 2018
-
-
Corey Farrell authored
This officially deprecates chan_sip in Asterisk 17+. A warning is printed upon startup or module load to tell users that they should consider migrating. chan_sip is still built by default but the default modules.conf skips loading it at startup. Very important to note we are not scheduling a time where chan_sip will be removed. The goal of this change is to accurately inform end users of the current state of chan_sip and encourage movement to the fully supported chan_pjsip. Change-Id: Icebd8848f63feab94ef882d36b2e99d73155af93
-
Corey Farrell authored
Add 'Section:' headings and use '-' for bullet points. Change-Id: I7e2be35601ac8fea53b90d926da564512b6716e4
-
Joshua Colp authored
-
Joshua Colp authored
-
Joshua Colp authored
-
George Joseph authored
-
George Joseph authored
-
Joshua Colp authored
-