* You are viewing Posts Tagged ‘asterisk’

Changing To The Linear Strategy Currently Requires Asterisk To Be Restarted

Hello,

using asterisk 1.8.12.2 and realtime architecture with mysql.

I get the following message on CLI when changing the value in the strategy

/[Mar 26 11:02:24] WARNING[10648]: app_queue.c:2030 queue_set_param:
Changing to the linear strategy currently requires asterisk to be restarted.//
//[Mar 26 11:02:24] WARNING[10648]: app_queue.c:2030 queue_set_param:
Changing to the linear strategy currently requires asterisk to be restarted./

Can this be done without restarting asterisk ?

Is this also the case in higher Asterisk versions ? For example Asterisk
1.8.24 ?




Kind regards, Jonas.

Asterisk 12.1.1 – Having Trouble Setting Up PJSIP

I am trying to make PJSIP work with my Cisco SPA504G phone. I have no problems making it work with the chan_sip driver.

When I configure my phone, it indicates the contact was added
– Added contact ‘sip:7001@192.168.9.142:5063′ to AOR ’7001′ with expiration of 3600 seconds

Phone shows green light for the line.

I then attempt to dial extension 1 and Asterisk crashes. I’m not seeing anything in the messages log.

I’m sure I’m doing something wrong, just not sure where to look or how to track down the problem. Can anyone offer some hints?

——————-

Skip ./configure When Source Directory Has Not Changed

Hi,

A silly question bouncing in my head for a long time :
when I’m installing-configuring a new Asterisk system, I’m using a script that issue the usual ./configure, make and make install commands to install Asterisk from source.

When installation fails for any reason, I would re-run my installation script which in turn, among many things, would launch the above ./configure command.

Is there a smart way to accelerate things a bit and skip ./configure when source files have not changed since last configure command was previously run ?

Regards

Getting T.38 Issue

Hi,

Few months back I configured Asterisk 11.6.0 for an outbound fax using T.38
protocol as listing down the flow below;

Asterisk Fax server -> (IP) -> Cisco VGW ->(IP) -> Carrier

The issue I’m currently getting when Asterisk receives warnings as listed below, it is overloading the Cisco VGW, therefore need to restart Asterisk service or sometimes reboot VGW to clear these warnings.

[Mar 24 09:25:01] WARNING[28645][C-00000004] app_fax.c: Unable to write frame to channel; Resource temporarily unavailable
[Mar 24 09:25:01] WARNING[28613][C-00000002] app_fax.c: Unable to write frame to channel; Resource temporarily unavailable

The configuration in sip.conf for T.38 is listed below;

t38pt_udptl = yes,fec,maxdatagram@0
faxdetect = t38

Udptl.conf;

udptlstart@00
udptlendI99
udptlfecentries = 3
udptlfecspan = 3
use_even_ports = no

Please advise at earliest to overcome this issue.

Need More Meetme Users — Hitting Some Limit

I’m trying to determine the capacity of my host running Asterisk 11.8.1 on CentOS 6.5.

The host is an Intel E3-1240v3 with 8GB RAM, an SSD, and gigabit Ethernet.

The primary application will be bridging groups of users using meetme().

I’m using 2 boxes — 1 to initiate calls using call files (box1), and 1
behaving a bit more like a production box — bridging calls (box2).

The call file on box1 originates a call to box2 and then plays a 2 hour WAV file.

The dialplan on box2 drops the call into a meetme, creating the room name from the last 2 digits of the current call count — distributing the calls into 100 meetmes.

When I run a script to create 500 call files on box1, box2 starts complaining at 312 calls, logging ‘Unable to open DAHDI pseudo channel:
Cannot allocate memory’ on the console.

From the ‘callers perspective’ the call is dropped between ‘There are currently x other participants in the conference’ and the ‘beep-beep.’

‘top’ says Asterisk is only using about 1/2 gigabyte of RAM.

‘top’ says Asterisk is using about 250% of the CPU (4 physical, 8 logical cores).

‘ulimit’ (added to /usr/sbin/safe_asterisk in the run_asterisk() function)
says the open file limit is 397,006.

‘ls -l /proc/$(cat /var/run/asterisk/asterisk.pid)/fd | wc -l’ says Asterisk only has 2,194 files open.

‘iftop’ sees about 24Mb of bandwidth in each direction between the boxes.

Using confbridge() I can easily get 3,000 calls (14,869 open files, 180Mb bandwidth), but I’d lose some functionality and have to re-write parts of my application.

Any clues of what limit I’m hitting and how to increase it?

XMPP Issues In Asterisk 11.6.0 For Distributed Device States…

I have been working with distributed device states in Asterisk using XMPP
attached to an OpenFire server. I have it working well across two servers and want to roll it out across every server in my company. All servers are Asterisk 11.6.0. I am running into a problem that seems like it should be a bit easier to solve than it is seeming to be. On the third server I am rolling into this solution, I get plenty of the following:

res_xmpp.c:1398 xmpp_pubsub_handle_error: Error performing operation on PubSub node device_state, 403.

So, basically, servers 1 and 2 continue to hum along nicely updating their device state, but server 3 gets a 403 forbidden message when it tries to deal with device state. I believe this has to do with the permissions set up on the device state node. I have a small example that demonstrates the creation of a new node.

In the Asterisk CLI, I ran ‘xmpp create collection asterisk test’ on server 3, which was successful and can be seen on servers 1 and 2 with
‘xmpp list nodes asterisk’

The debug output from server 3 for this is as follows:

<--- XMPP sent to 'asterisk' --->





http://jabber.org/protocol/pubsub#owner


collection


http://jabber.org/protocol/pubsub#node_config


1


1


whitelist



<------------->

<--- XMPP sent to 'asterisk' --->







<------------->

As we can see, the first message creates the test node and sets the access model to whitelist, so only jids in the whitelist are allowed to modify it. The second message then sets the appropriate server 1 and server 2
jids to be owners, thus meeting the requirements of the whitelist.

Since these nodes are persistent, it would appear that server 3 cannot properly access device_state because it was never whitelisted when the node was created originally. I am fairly certain that I can solve this by deleting all my nodes and letting them be recreated, but that seems extreme as I put more servers into the system. Any thoughts on a better way to handle xmpp and making sure new servers can access the proper nodes?


Kevin Larsen – Systems Analyst – Pioneer Balloon Company