MiContact Center v8 - Live Chat - Connection appears to be slowing down

[acloghesy]

Hello All,

We've recently implemented the MiCollab chat at our office. (we have the emails working for a while)
During testing all was well, however now that it's live we keep getting the following issues messages that are shown no our side in Ignite and the client sees them as well.
This is happening when there is more than one concurrent chat.
 
"Your connection appears to be slowing down. You may not receive messages from the agent at the time the agent sends them. If the problem persists you may want to check if you are running any other software that is using your internet connection."
Then after some time
"The connection to the server has been restored."

We have the MiContact center server with the following specifications.

Mitel MiContact Center Version 8.0
OS Name:                   Microsoft Windows Server 2008 R2 Standard
System Manufacturer:       VMware, Inc.
System Model:              VMware Virtual Platform
System Type:               x64-based PC

Processor(s):              2 Processor(s) Installed.
                           [01]: Intel64 Family 6 Model 44 Stepping 2 GenuineIntel ~2394 Mhz
                           [02]: Intel64 Family 6 Model 44 Stepping 2 GenuineIntel ~2394 Mhz

Total Physical Memory:     24,576 MB
Available Physical Memory: 8,233 MB
Virtual Memory: Max Size:  49,149 MB
Virtual Memory: Available: 32,827 MB


Any feedback is welcome. But help would be most appreciated!

[harsha2690]

Hello acloghesy,

 I cannot determine exactly what's happening. But can provide u with few steps of troubleshooting
1, Check if you have any Alarms in MiCC . If so, Try to find a Knowledge base article for that Alarm.
2, Check if this is happening even after disabling firewall on server and client.
3, Is this is restricted to certain agents ? If so, how are they different .

I hope this steps will help you to narrow down your issue .

Thanks .

[acloghesy]

HI Harsha
Thanks for your response.
I looked through the Alarms again, nothing related or at least standing out.
We have made no changes and the oddest thing is that this happens sporadically through only a few chats throughout the day.
I verified there are no heavy scheduled tasks on the server or on the SQL server.

And we only have one agent that takes car of the chat, since we only recently implemented it.
I verified her network adapter and everything hardware/software wise on her PC is good.

Thanks,
AC

[acloghesy]

These are the only errors we are seeing.
Attached are the screenshot on the Agents side.
also the errors that I did see in the PrairieFyre logs.

[pFoldtimer]

The reason you are seeing this is because the underlying signalR connection detects a slow connection or the connection is going up or down.
The log you want to look at is the ccmwa log in the logs for on the server to check for anything unusual as well as hitting F12 in Chrome and looking at the console logs. It would be interesting to see if the signalR connection transport is using websockets or server sent events or long polling.
Do you see this only on the customer side of the chat or on the agent side as well?
Some customers end up having proxies or firewalls that interfere with the signalR connection.
Did you enable/install websockets within IIS on your sever?

[acloghesy]

Hi pFoldtimer,
Thanks for the response, unfortunately we are hosting the Ignite server on 2008, so no Websockets.
I will look deeper into the logs and post my findings.
The messages are seen on both the Client & Agent side, and it's not persistent. it tends to happen every 5-6 chats.
Thanks,
AC