Hello, I have a question about the "priority" in the dispatcher module. I looked in the docu-wiki and there was nothing which actually described how this field is used (exactly), so I can only guess or ask the list.
Say I have 3 destinations in a set:
set: 1 dest: sip:1.2.3.4:5060 priority: 1
set: 1 dest: sip:1.2.3.5:5060 priority: 1
set: 1 dest: sip:1.2.3.6:5060 priority: 0
I am using ds_select_domain("1","4") to retrieve my list of destinations.
In this scenario, would the calls only route to 1.2.3.6 if .4 and .5 were both out of service?
Thanks.
Hello,
On 8/26/10 3:23 PM, Geoffrey Mina wrote:
the priority is used to sort the records in memory. IIRC, the highest priority the first in the list (for same priority should be the order returned by database query result).
If you use round-robin, the the first destination used is the one with highest priority, the next, ... and loop again.
The priority is particularly good for classic serial forking (algorithm 9), where always first destination is the highest priority.
Cheers, Daniel
OK, so In my scenario this wouldn't achieve what I am looking for. I basically want 2 servers to get all the round-robin requests unless they are unavailable. Then I would want the lower proiority servers brought into the mix.
Maybe ill take a look at the source to see how tough this would be.
Thanks.
On 8/26/10, Geoffrey Mina geoffreymina@gmail.com wrote:
You can just do this yourself with script arrays and custom database queries. Easily. Why is everyone so hung up on trying to squeeze logic functionality from modules that isn't there?
On 08/26/2010 10:47 AM, Geoffrey Mina wrote:
On 8/26/10 4:47 PM, Geoffrey Mina wrote:
you can do that. It is up to your logic in config -- use as may setids as you need in cascade. E.g., just build the logic so that if all routes in dispatcher setid X fails then use routes from setid Y.
Cheers, Daniel
-
Alex Balashov -
Daniel-Constantin Mierla -
Geoffrey Mina