Topics

1

Running QueueMetrics / Cluster, agent logon page problem. Asterisk 1.8, shows SIP not Local (Resolved)

« on: September 20, 2017, 16:10:32 »

The last post has been closed. So I'm opening this new "Resolved" post to help any others having this same problem and for our use down the road if needed.

So here is the rest of the story.

First,,,, here is our last post on this subject before it was closed:

************************************************************************************

Re: Cluster, agent logon page problem. Remote Asterisk 1.8, shown as SIP not Local
« Reply #16 on: July 18, 2017, 08:17:53 »
We added this bug to our bug tracking system and our developers will solve it accordingly to their schedule.
Thanks for your understanding.

************************************************************************************

Here is the Resolution to Resolve this issue:

After two months of going back and forth on this we found that the clock in the Cluster and the Remote Asterisk Server where 2.5 minutes off from each other.

WOW........ once we changed the clocks to be the same, then adding NTP on the Cluster everything was working.

The problem we had and that is now revolved, was when using Cluster, Web page, Remote Agent screen wouldn't update as it should. Unless you updated your WEB page then it would show/update correctly.

What a happy day when we saw the Agent web page update on it's own. Now the screen updates as it should......

It's amazing how something so easy/simple can be over looked while using, exercising your best intentions and judgement for a repair of a problem.

I didn't follow my own advice that's given often "DUMB IT DOWN"........... Often if we are like Plankton at the top of the sea and not on the bottom with the sea cucumbers you will find the issue/problem much sooner. Don't dive so deep as we technical people like to do. Thinking we have a real problem here and are going to save the world using our smartness. Most often the easiest and quickest way to resolve an issue/problem is to dumb it down.

Hats off to everyone involved for there efforts, Loway, users and our staff.

What a great product, work and staff members that comes out of Loway.

/rk

2

Running QueueMetrics / QueueMetrics Cluster, Agent WEB Page Source file

« on: June 26, 2017, 16:09:23 »

QueueMetrics Cluster, Agent WEB Page Source file

Here is the situation:

We/you have a single dedicated QueueMetrics server in QM Cluster Mode. It’s running nothing but Linux and QueueMetrics (QM) - 16.10.13
Then you have several other remote servers (sites) running Linux, 1.8 Asterisk, FreePBX and QM - 12.5.1 on each of them.

Here is the Question(s):

If you go into the QM Cluster server. Then login to the “Agent’s Web Page” and Log an agent into a Queue, where is QM running the AMI command from?
Is the AMI commands to accomplish the Logging on/out/Pause/Unpause coming directly from the QM Cluster?
Or is QM Cluster Server, telling the remote QM Server not running Cluster, to do the AMI commands with the /etc/asterisk/extensions_queuemetrics.conf file?
I’m trying to figure which file the QM Cluster is using to send the AMI Logging on/out/Pause/Unpause from.

Here is the problem:

We are having a problem when using QM Cluster Software logging on Agents as SIP/XXXX into a Queue in the remote server. Instead of logging on agent(s) using Local/XXXX this remote servers Queue. Changing this SIP to Local in the QM Cluster Web Agent page (where ever it’s doing the AMI from) should make the Logon/off/Pause/Unpause work. This would make it work when we have a SIP phone not being seen by the system, plugged into the network or registered/attached. Along with the, Follow Me and Dial modification. All of this for call out on the PSTN line(s) for remote queue agents, on a SIP configured extension to work properly.

Here is what we know, think and wish for:

QM recently released and said:
"install of QM using the latest 17.06.1. After this, check for DirectAMI options and change from SIP/${num} to Local/${num}@from-internal/n"

So QM being the great company that they are, is seeing the need and have a fixed for this, as far as we know. We have not installed 17.06.1 yet.

We feel, see and have found that from all indications thus far, that the QM Cluster Server is sending out these AMI commands directly to the remote QM server to Logon/off/pause/unpause agents.

So where is this QM Cluster file? Where does the AMI commands originate from? So one could see, learn and possibly change this file if one desired to?

Thank you QueueMetrics for a great product.

/rk

3

Running QueueMetrics / Cluster, agent logon page problem. Remote Asterisk 1.8, shown as SIP not Local

« on: June 22, 2017, 23:48:49 »

Cluster, agent logon page problem. Remote Asterisk 1.8, shown as SIP not Local

We have 55 Asterisk 1.8 Asterisk servers with thousands of phone. Been using QM for years and love it…..

What we have done is built a SIP phone (not registered/hooked up to anything) in the 1.8 Asterisk as a sip extension XXXX having the "Dial" set to Local/9XXXXXXXXXX@from-internal/n. This “Dial” is in the FreePBX SIP extension build page. This 9XXXXXXXXXX is so when you dial or call SIP XXXX it will call out PSTN to a remote queue agent. We have also built this XXXX extension into our queue, as a dynamic agent.

How we login into our QM Cluster machine using this agent/XXXX. Once logged in we make this agent XXXX, logged into our queue. Note: We are counting on the pause/up-pause in the QM Cluster agent page to work. As remote agents need this feature.

Here is the issue:
When running QM Cluster using the agent web page you log into the QM agent web page by using your agent/XXXX. Once logged in, you log your agent into the queue. The queue they are already dynamically assigned too. It all seems to work fine to this point.

Here is the problem.

You look into your remote 1.8 Asterisk you type in: asterisk -rvx 'queue show XXXX'. It shows your extension is logged into your queue as "SIP/XXXX" not "Local/XXXX".

Here is what it looks like (other than the XXXX) using the QM Cluster:
   Members:
      SIP/XXXX (dynamic) (Not in use) has taken no calls yet

Here is what it should look like (other than the XXXX) if you are not using QM Cluster:
Members:
      XXXX Some Queue (Local/XXXX@from-queue/n) (dynamic) (Not in use) has taken no calls yet

We feel this SIP/XXXX vs Local/XXXX is our problem when running 1.8 Asterisk/ FreePBX 2.9.0.12 with QM.

In your 1.8 Asterisk you can build a SIP extension XXXX with a Dial: Local/9XXXXXXXXXX@from-internal/n
This Local/9XXXXXXXXXX is so our queue calls will go out to remote agent(s) cell/Home/office across our PSTN lines. The problem is when you call the queue number, it doesn’t know your agent is logon because it’s logon as an SIP/XXXX not a Local/XXXX.

As soon as you logout your agent from the QM Cluster agent page. Then log back in using any other phone/softphone, even when using the toggle feature, *45XXXX*XXXX it logs your back correctly as shown below (other than the XXXX and Some Queue description).

XXXX Some Queue (Local/XXXX@from-queue/n) (dynamic) (Not in use) has taken no calls yet

If you try calling your queue using any logged in phone using *45XXXX*XXXX it works.

But if you logout and back in using QM Cluster agent page, it breaks it. Because 1.8 Asterisk doesn’t seem to want to see QM Cluster's SIP/XXXX. But wishes to see Local/XXXX………..

Here is the version and equipment list of the TWO machines. QM Cluster and the Asterisk/FreePBX/QM remote machine:

QM Custer is a dedicated machine shown here:

Software release:   Loway QueueMetrics - 16.10.13
B: 184 - 2017.03.23-08:34
Firefox client version:   CURRENT - Install now: Firefox – Chrome
Operating System:   O.S.: Linux - Ver: 3.10.0-514.el7.x86_64 - amd64
Java Runtime:   Version: 1.6.0_22
Vendor: Sun Microsystems Inc.
Class Version: 50.0
Java Home: /usr/local/queuemetrics/jdk1.6.0_22/jre
System path:   /usr/local/queuemetrics/tomcat/webapps/queuemetrics/WEB-INF
Storage type:   Cluster (Members: *)
Loway TPF:   Version: TPF 264p/P B:264p - 2016.11.29-16:11
Language pack:   V:2017-03-16 10:39
System time:   Java Time: 2017-06-22 13:25:05
Java Time Zone: America/Boise
MySQL Time: 2017-06-22 13:25:05.0
MySQL Time is aligned with Java Time

Here is the 1.8 Asterisk/FreePBX/QM Remote machine:

Software release:   Loway QueueMetrics - 12.5.1
B: 4015 2012-05-15 21:37
Firefox client version:   0.2.6 - Install now: Firefox - Chrome
 
 
Operating System:   O.S.: Linux - Ver: 2.6.32-220.17.1.el6.i686 - i386
Java Runtime:   Version: 1.6.0_22
Vendor: Sun Microsystems Inc.
Class Version: 50.0
Java Home: /usr/local/queuemetrics/jdk1.6.0_22/jre
System path:   /usr/local/queuemetrics/tomcat/webapps/queuemetrics/WEB-INF
Storage type:   Flat file: (/var/log/asterisk/queue_log)
Loway TPF:   Version: $Revision: 1.93 $ J:193/P B:25
Language pack:   V:2012-05-09 15:24
System time:   Java Time: 2017-06-22 13:28:59
Java Time Zone: America/Boise
MySQL Time: 2017-06-22 13:28:59.0
MySQL Time is aligned with Java Time

Thoughts? Help? Suggestion? Corrections?

/rk