Summary: | ASTERISK-00124: RingAall strategy, using mix of agentlogin, agentcallbacklogin does not ring first avaiable after retry time expires | ||
Reporter: | tclark (tclark) | Labels: | |
Date Opened: | 2003-08-19 01:43:37 | Date Closed: | 2004-09-25 02:14:14 |
Priority: | Minor | Regression? | No |
Status: | Closed/Complete | Components: | Applications/app_queue |
Versions: | Frequency of Occurrence | ||
Related Issues: | |||
Environment: | Attachments: | ||
Description: | ( This happens when I am not getting random mem corruptions) 1) set up a mix'd q using agentlogin, agentcallbacklogin 2) Login both types of agents 3) stack 3 calls into the q 4) Call 1 goes to agentlogin channel 5) call 2 goes to agentcallbacklogin 6) hang up call 2 on the agentcallbacklogin 7) the thrid call will now wait 30secs untill it starts to ring the agentcallbackchannel, it should ring any free agent after the retry time, *NOT* the timeout period ****** ADDITIONAL INFORMATION ****** queues.conf [someq] strategy = ringall timeout = 30 retry = 1 maxlen = 0 member => Agent/:2 ;first avaiable in agent group 2 agents.conf musiconhold=>none group=1 ackcall=no agent => 100,,DispatchOP group=2 ackcall=no agent => 101,,OrderEntryOp extensions.conf [dispatchlogin] exten => *1,1,AgentLogin(101) ; exten => *3,1,AgentCallbackLogin(100|@dispatchlogin) ; exten => 100,1,Dial(Zap/1,20) ;dial ext on zap channel 1 exten => 100,2,Congestion | ||
Comments: | By: Mark Spencer (markster) 2003-08-19 09:58:47 I've added some experimental code to make callback agents show back up. I suggest you add Agent/:1 in your example, so that it will ring both sets of agents. By: tclark (tclark) 2003-08-21 16:44:52 Shall we close this one as technically impossible ??? By: Mark Spencer (markster) 2003-08-22 09:53:58 This behavior is not reasonable to implement. |