[Asterisk-Users] Call Queue NOT using RoundRobin ?!?

Michael Konietzny michael at konietzny.ws
Thu Jun 29 12:44:07 MST 2006


hey,

a patch for linear mode is posted to bugs.digium.com already:

http://bugs.digium.com/view.php?id=7279

greetings,
 Michael

Aaron Paxson schrieb:
> If someone can point me in the right direction, I'll look into it. 
> I'm not a C programmer, but I *should* be able to find my way.
>  
> I'm looking at app_queue.c  I see the strategies defined, but nothing
> about how they are used.  Is app_queue.c the file that does the calling?
>
>     ----- Original Message -----
>     *From:* Alessio Focardi <mailto:afocardi at gmail.com>
>     *To:* Asterisk Users Mailing List - Non-Commercial Discussion
>     <mailto:asterisk-users at lists.digium.com>
>     *Cc:* aj at thepaxson5.org <mailto:aj at thepaxson5.org>
>     *Sent:* Thursday, June 29, 2006 2:07 PM
>     *Subject:* Re: [Asterisk-Users] Call Queue NOT using RoundRobin ?!?
>
>     Will you (or anyone else) be able to code this proposed "circular"
>     or "linear" (what sounds more appropriate?) strategy and submit it
>     for inclusion in HEAD ?
>
>     Should be pretty easy, unfortunately I have very few programming
>     skills.
>
>     Regards !
>
>
>     P.S.
>
>     here is a snippet from the wiki, whatever it means ! :)
>
>     roundrobin mode remembers the last agent it _started_ with for a
>     new call, and starts with the next agent in the list. If you have
>     three agents, the first call will go to agent 1->2->3, the next
>     call will go to 2->3->1, the next call will go to 3->2->1, etc.
>
>     rrmemory mode remembers the last agent it tried to _call_,
>     regardless of who it started with, so that the next call will go
>     the agent after the last one who answered. If you have three
>     agents and the first call rings 1->2 (and is answered), then the
>     next call will ring 3->1 (and is answered), then the next call
>     will ring 2->3->1, etc. For the first call, if agent 2 answered it
>     in roundrobin mode, they would still be the first agent for the
>     next call, but rrmemory mode will move past them.
>
>
>     On 6/29/06, *Aaron Paxson* <aj at thepaxson5.org
>     <mailto:aj at thepaxson5.org>> wrote:
>
>         The linear function helps me too.  I've built an extensive
>         multi-queue technical support system strategy.  Based on the
>         initial queue, ALL calls goes to Tier1 first.  Then, if Tier1
>         does not get the call (on the phone/away from desk), Tier2
>         should get it, so on, and so forth.
>          
>         In Tier1, the primary helpdesk technician (like your
>         receptionist idea) takes ALL calls (That's what they were
>         hired for).  However, others can help out, if the pri
>         technician is on the phone.
>          
>         Here's my question:
>          
>         If roundrobin strategy remembers the last call made, and sends
>         the next call to the next number (and this is by design), then
>         why on earth was the RRMemory strategy created??
>          
>         Thanks for your response, Alessio.
>          
>         ~~Aaron
>
>             ----- Original Message -----
>             *From:* Alessio Focardi <mailto:afocardi at gmail.com>
>             *To:* Asterisk Users Mailing List - Non-Commercial
>             Discussion <mailto:asterisk-users at lists.digium.com>
>             *Cc:* aj at thepaxson5.org <mailto:aj at thepaxson5.org>
>             *Sent:* Thursday, June 29, 2006 1:31 PM
>             *Subject:* Re: [Asterisk-Users] Call Queue NOT using
>             RoundRobin ?!?
>
>             Welcome to my personal hell ! :)
>
>             I'have been discussing this previously on the list and
>             also with some digium staff: to my experience there is NO
>             way to archieve a linear distribution of calls from a queue.
>
>             I mean
>
>             When a call comes in first member of the queue is ring,
>             then second, etc
>
>             Subsequent calls take the same path: first, second and so on.
>
>             Someone has suggested to use "ringall" with penalties
>             (pretty esotic!) but also this is not working for the
>             purpose.
>
>             I was also told that "nobody wants that" (you insensitive
>             clod!) even if this call distribution seems pretty logic
>             in some case scenarios.
>
>             (hint: a receptionist is first member of a queue and
>             another person is the second ... receptionist goes for a
>             pee and magically calls are rerouted to the backup
>             operator after ringing to the first).
>
>             Hope you can find out something to share, maybe we can
>             also launch a "count us" initiative :)
>
>             Alessio Focardi
>
>
>
>
>             On 6/29/06, *Aaron Paxson* <aj at thepaxson5.org
>             <mailto:aj at thepaxson5.org>> wrote:
>
>                 I have setup several Calling Queues, each setup with
>                 RoundRobin strategy.   When I call the queue, the
>                 first member/agent phone rings.  Great!  I call it
>                 again, the second member/agent rings??
>                  
>                 I thought that was the RRMemory strategy, but it seems
>                 RoundRobin is also doing it.
>                  
>                 Anyone know what I can do to my queues, in order to
>                 force each call down the ordering of my members list?
>
>                 _______________________________________________
>                 --Bandwidth and Colocation provided by Easynews.com
>                 <http://Easynews.com> --
>
>                 Asterisk-Users mailing list
>                 To UNSUBSCRIBE or update options visit:
>                   http://lists.digium.com/mailman/listinfo/asterisk-users
>
>
>
>
>         _______________________________________________
>         --Bandwidth and Colocation provided by Easynews.com
>         <http://Easynews.com> --
>
>         Asterisk-Users mailing list
>         To UNSUBSCRIBE or update options visit:
>           http://lists.digium.com/mailman/listinfo/asterisk-users
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> --Bandwidth and Colocation provided by Easynews.com --
>
> Asterisk-Users mailing list
> To UNSUBSCRIBE or update options visit:
>    http://lists.digium.com/mailman/listinfo/asterisk-users
>   




More information about the asterisk-users mailing list