FIPPA ready to not ready state for all users randomly every 2-30 mins

[pinfat1987]

Hello Team, 

We have a cluster of UCCX and CUCM in our infra and we recently upgraded UCCX to 11.6.2. 

Some particular department are experiencing issues with their desk phone through both FIPPA and Finesse. 

When i check login-logout report, i see connection failure. 

All users in that department have their phone state from ready to not ready randomly, it can be every 2 mins or 30mins or hours. 

I performed ping test from UCCX to phone and there are no packet loss. 

The issue is only with that particular deparment. Also, phone model is 7942. 

Can you please help or advise on what to focus to fix this type of issues? 

Thank you. 

[richard-ferguson]

Having same issue with same conditions: CCX 11.6.2 (Standard), 7942, Extension Mobility with Connection Failure after 2 minutes.

We use FIPPA for these remote Agents and due to Standard CCX have no other option, so don't suggest using a browser as it's not possible with Standard CCX (Standard should be outlawed eh).

Agents go: LOGIN --> READY (wait 2 minutes) --> LOGOUT with Connection failure. The CUIC Agent State Detail report and logs both indicate that NOT READY is being skipped completely, which is not possible if you're an Agent. READ FLAG.

Although Cisco state 7942 is NOT SUPPORTED they work fine on other CSQs on the same floor at the same time so I'm putting that aside as a contributor unless Cisco TAC can prove otherwise.

It's a new queue. The old queues are working fine. CUCM and CCX configis have been re-created from scratch using our in-house Agent On-Boarding guide that ensures consistency for this new CSQ compared to the reliable old CSQs.

The CCX Trigger CSS lists the Agent PT so it's unlikely to be that.

It's not the System Parameter "Change Agent State to Not Ready when Agent Busy on Non ACD Line" because a/. it's 'Disable' and b/. they haven't received any calls yet anyway.

Agents in another city, adjacent to the CCX server stay logged in AOK. So it looks like network or location-related issue but the web server on the phones reports low latency (0-30ms) and 0 dropped packets.

Stumped ATM. Any suggestions?

[abdallah.s69@yahoo.com]

you could try "set webapp session timeout [minutes]"

[richard-ferguson]

Hey thanks for that tip - two questions before I follow up:

1/. Does it apply to FIPPA Agents?

2/. Would it affect the other 60 Agents who are not affected? I don;t want to unleash chaos on happy workers :-)

[pinfat1987]

Thanks for the suggestion Abdallah !!
The fix was as per below:
utils finesse ippa_inactivity_timeout enable 86400
and restart of finesse service.

After 2 months of troubleshooting :) cisco tac finally concluded that it was due to a bug!!!

[silvakizir]

Hi.

Can you share with us the Bug id?

[pinfat1987]

Hi Richard Ferguson, its related to a bug and it was fixed long time back :)
. configure below:
- utils finesse ippa_inactivity_timeout enable 86400
- restart finesse service.

Regards,
Pin

[richard-ferguson]

Cool thanks team.

Have applied that before the start of the business day and will monitor today and post an update when I have one.

I note for others who read this that the CLI command offers to restart 'Cisco Finesse Tomcat' service automatically - very  convenient as this is normally only do-able from the admin WebUI. Was back in service in under 10 minutes.

[richard-ferguson]

This is an odd feature - it seems to take several days to become active on handsets and when it does, it doesn't seem to stick to the nominated number of seconds of inactivity (might do, but it's not obvious). It has started to auto logout badly taught Agents who stay logged in ad-infinitum and we're hoping it also logs out zombie Agent logins that happen for home-agents.

Thanks Community people :-)