Genesys Cloud - Main

 View Only

Sign Up

  • 1.  Not Responding Status

    Posted 20 days ago

    Hello everyone.

    I am having an issue where agents get placed on the Not Responding status. Sometimes I manually switch the agent status on the dashboard until they receive a call but in some cases this doesn't work and so we have the following workaround:

    1. Log out the agent from Genesys.
    2. Delete cache and cookies from the browser
    3. Do a DNS cache flush
    4. log the agent from their users on the windows PC and log back in.

    Anybody else had something like this happen? Any tips?

    Our IT team is looking into this right now so I'll update this thread for the benefit of the community if we resolve this.


    #Omni-ChannelDesktop/UserInterface
    #Routing(ACD/IVR)

    ------------------------------
    Juan C Rodriguez Bracero
    ------------------------------


  • 2.  RE: Not Responding Status
    Best Answer

    Posted 20 days ago
    Hi Juan,
     
    I found a couple of similar discussions where the issue was ultimately related to connectivity rather than the Not Responding status itself.
     
    One additional thing worth checking is whether the affected agents have multiple active sessions. For example:
     
    Multiple browser tabs open
    Desktop app and browser running simultaneously
    The same user logged in from different devices
     
    Multiple concurrent sessions can lead to missed notifications and inconsistent behavior if they're not part of a supported configuration.
     
    If the affected users are using WebRTC, another useful step is running the built-in WebRTC Diagnostics (Calls → Phone Settings → Run Diagnostics).
     
    The diagnostics provide metrics such as packet loss, latency, jitter, and MOS, which can help identify underlying network issues.
     
     
    If the issue continues after verifying the agent sessions and reviewing the diagnostics, collecting HAR logs and browser console logs would be the next step.
     
    If the issue still cannot be isolated, it may be worth opening a Customer Care case so the support team can review the backend logs and investigate further:
     
    https://genesys.my.site.com/customercare/GenesysCommunityLogin
     
    It's great that your IT team is already investigating. Hopefully the WebRTC diagnostics and checking for multiple active sessions will help identify the root cause.
     
    Please keep us posted on what you find. It could be very helpful for others experiencing the same behavior.


    ------------------------------
    Raphael Poliesi
    ------------------------------



  • 3.  RE: Not Responding Status

    Posted 9 days ago

    Hello everyone,

    Thanks for the replies.

    There are some things on the initial post that I didn't explain well.

    We have all our queues set to auto answer. Sometimes an agent will get a call that instead of answering automatically, it will "ring" and users have the choice of answering or declining. Clicking on answer doesn't work and after 8 seconds or so of ringing, the agent gets placed on the Not Responding status. Whenever this happens, I have the user go into phone settings and run diagnostics. The "Able to reach Genesys services via DNS" test fails. To solve this, I clear the browser cache, log out the user from Genesys, clear the DNS cache and then i have the user log out and back in on the computer.

    What our IT team has found so far is that the user's computer may lose communication randomly to our domain controller and that triggers the issue. We have been monitoring this the last few days, so I'll let you guys know.

    Thanks.



    ------------------------------
    Juan C Rodriguez Bracero
    UNIVERSAL INSURANCE
    ------------------------------



  • 4.  RE: Not Responding Status

    Posted 20 days ago

    Hi Juan,

    Just to add to what Raphael has already said, we've seen similar behaviour in our environment as well. While we continued investigating the root cause, we implemented a workflow that detects agents entering the Not Responding state, automatically returns them to Idle, and sends a Slack notification to our support team. It doesn't address the underlying cause, but it has helped minimise the operational impact and reduced the need for manual intervention.

    In our case, we know Not Responding can occur if an agent doesn't answer an offered interaction, but we've also seen it occur unexpectedly due to what appeared to be connectivity or session-related issues. I'd be interested to hear what your IT team identifies as the root cause, so please keep us posted.



    ------------------------------
    Phaneendra
    Technical Solutions Consultant
    ------------------------------



  • 5.  RE: Not Responding Status

    Posted 11 days ago

    Hi Phaneendra,

    Can you guide me with the workflow implemented that detects agents entering the Not Responding state, automatically returns them to Idle, and sends a Slack notification to our support team. This is quite usefult.

    I do have a situation where we have On Premise edge for a country and due to regulatory constraints, we have restricted the Site communication to other Countries site which are not using On Premise edge.

    Scenario :

    Agent from another country Site dials the DID that is mapped to a call flow. Since this DID is configured within the same Genesys Cloud organization, the system treats this as an internal call. Genesys attempts to route the call internally from Cloud to On Premise Edge. Since, Due to regulatory restrictions, internal connectivity is blocked, and the call cannot be connected to the agent
    Because the call cannot be completed, Genesys still assumes the destination is reachable. The call is repeatedly offered to agents in the queue. Each failed attempt results in agent being marked as Not Responding. This process continues until all available agents in the queue are impacted and moved to Not Ready.

    We are deploying some workaround but, in the meantime, your workflow can help me.



    ------------------------------
    Arpit Arora
    ------------------------------



  • 6.  RE: Not Responding Status

    Posted 11 days ago
    Hi Arpit
     
    Happy to share the approach we used.
     
    Trigger
    Topic: v2.users.{id}.activity
    Trigger Condition (JSON Path):
    routingStatus.status == "NOT_RESPONDING"

    Workflow

    1. Extract the required values from the trigger payload. For example:

      ToString(GetJsonObjectProperty(Flow.jsonData, "id"))

      This gives you the userId. We also extract the routing status.

    2. Retrieve the user's details:
      GET /api/v2/users/${input.userId}

      We use this to obtain the agent's name and division. If you only want this automation to run for specific divisions, you can perform a division check here.
    3. If the division matches (or if you don't need a division filter), reset the agent's routing status:

      PUT /api/v2/users/${input.userId}/routingstatus

    4. Finally, send a notification. We use the Send Notification action to email a monitored mailbox that automatically forwards to a Slack channel. Alternatively, you could send directly to Slack using an Incoming Webhook.

      This has worked well for us in automatically recovering agents from the Not Responding state while also notifying the support team so recurring issues can be investigated.

      In your scenario, it should help minimise the operational impact while you work on the underlying Cloud ↔ On-Premises Edge routing issue.

      Depending on how frequently the failed call offers occur, you may also want to add a short delay (for example, 5–10 seconds) before resetting the routing status. Another option is to implement a maximum number of automatic resets per agent. This can be achieved by tracking the reset count in a Data Table. Once the threshold is reached, stop automatically resetting the agent and instead send an alert for manual investigation. You could also have a scheduled process clear the reset counts periodically (for example, daily or hourly), allowing the automation to resume. This helps avoid repeatedly cycling an agent between Not Responding and Idle if the underlying issue persists.

      Hope this helps.


    ------------------------------
    Phaneendra
    Technical Solutions Consultant
    ------------------------------



  • 7.  RE: Not Responding Status

    Posted 10 days ago

    Thanks Phaneendra, I will definitely try this.



    ------------------------------
    Arpit Arora
    ------------------------------



  • 8.  RE: Not Responding Status

    Posted 10 days ago

    If you need any help just let me know, thank you.



    ------------------------------
    Phaneendra
    Technical Solutions Consultant
    ------------------------------



  • 9.  RE: Not Responding Status

    Posted 9 days ago

    Hi Arpit, I would appreciate, if you can share your test result.



    ------------------------------
    Manish Suryavanshi
    Lead Consultant
    ------------------------------



  • 10.  RE: Not Responding Status

    Posted 7 days ago

    Discussão muito interessante.

    Juan, a última informação que você compartilhou me parece especialmente importante: vocês utilizam auto-answer, mas, quando o problema ocorre, a interação passa a tocar para o agente e o teste "Able to reach Genesys services via DNS" falha.

    Nesse cenário, eu trataria o Not Responding mais como uma consequência do problema do que como a causa principal. A investigação de conectividade/DNS que vocês estão realizando parece ser um caminho importante para encontrar a causa raiz.

    Também gostei bastante da automação apresentada pelo Phaneendra como forma de reduzir o impacto operacional enquanto a investigação continua. Eu apenas adicionaria alguns controles para evitar que a automação acabe mascarando um problema persistente:

    • limitar a quantidade de resets automáticos por agente dentro de uma janela de tempo;

    • registrar horário, usuário, site/divisão e quantidade de ocorrências;

    • após atingir um limite, parar de retornar automaticamente o agente para Idle e gerar um alerta para investigação;

    • correlacionar os horários das ocorrências com os diagnósticos WebRTC e logs da infraestrutura.

    Assim conseguimos separar bem duas responsabilidades:

    Mitigação: recuperar rapidamente a operação.

    Observabilidade: preservar informações suficientes para identificar e corrigir a causa raiz.

    Esse tipo de automação com Trigger + Workflow é um ótimo exemplo de como o Process Automation pode ajudar a reduzir impacto operacional sem substituir a investigação técnica do problema.

    Juan, seria muito interessante saber o resultado final da análise do seu time de infraestrutura quando vocês identificarem a causa.

    __________________________________________________________________________________________________________________________________________________________________

    Very interesting discussion.

    Juan, I think the latest detail you shared is particularly important: your queues use auto-answer, but when the issue occurs, the interaction starts ringing for the agent instead, and the "Able to reach Genesys services via DNS" diagnostic test fails.

    In this scenario, I would look at Not Responding more as a consequence of the underlying issue rather than the primary cause. The connectivity/DNS investigation your IT team is already performing seems like an important path toward identifying the root cause.

    I also really like the automation Phaneendra shared as a way to reduce the operational impact while the investigation continues. I would just add a few guardrails so the automation does not unintentionally hide a persistent problem:

    • limit the number of automatic resets per agent within a defined time window;

    • record the timestamp, user, site/division, and number of occurrences;

    • once a threshold is reached, stop automatically returning the agent to Idle and generate an alert for investigation;

    • correlate the occurrence timestamps with WebRTC diagnostics and infrastructure logs.

    This creates a useful separation between two responsibilities:

    Mitigation: quickly restore the operation.

    Observability: preserve enough information to identify and fix the root cause.

    This type of Trigger + Workflow automation is a great example of how Process Automation can help minimize operational impact without replacing the technical investigation of the underlying issue.

    Juan, it would be very interesting to hear the final outcome from your infrastructure team once the root cause is identified.



    ------------------------------
    Matheus Mendonca
    ------------------------------