Versions Compared

Version Old Version 8 New Version 9
Changes made by

Mack Baczynski

Mack Baczynski

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Info

One-way audio can be frustrating and difficult to solve because you only have controls of a few of the required components to make this work. When an agent cannot hear the remote side, in most cases it is either a speaker configuration issue, or a problem with the remote end. It’s recommended in these situations to make repeat attempts to see if the behavior is consistent.

Applicable guides

Call connects. Agent can hear remote side. Remote side cannot hear agent

...

Info

In most situations, this will be caused by a local microphone connection or configuration with the agent’s workstation. It’s always good practice to place controlled test calls to validate 2-way audio before the start of business if you’re concerned with one way audio.

Applicable guides

Unexpected dropped call

Potential Causes

Common Solutions

  • Remote side carrier service interruption

  • Remote side device fault

  • Agent ISP service interruption

  • Agent local network service interruption

  • Agent browser interruption

  • Agent workstation interruption

  • Agent audio device interruption

  • Sharpen voice infrastructure interruption

  • Re-engage with party for which dropped call occurred. Determine whether this was expected or unexpected on their end.

  • Inquire with agent’s ISP to determine whether there was a known outage at the time of the unexpected event.

  • Validate local network equipment uptime and performance statistics. Determine whether competing local network activity such as streaming/gaming/heavy downloading occurred at the same time.

  • Validate browser operation. Are web pages responsive or slow to load? Confirm the browser tab did not freeze or consume system resources.

  • Validate user workstation has available resources. CPU, Disk, or memory exhaustion which interrupt timely functions can cause our webRTC connection to drop. Adding memory or upgrading the workstation may be necessary.

  • Validate audio device (microphone/headset) did not experience a disconnect or fault. Sometimes, especially with usb connected devices, a brief disconnection can cause the OS and browser to loose knowledge of the device, and drop the call.

  • Work with SharpenCare to observe and troubleshoot the health of webRTC and Voice Processor infrastructure. We may ask to retrieve browser logs (if available) to more specifically identify the issue.

Info

Dropped calls, as the list above suggests, have a variety of potential root causes. Most dropped calls are a result of poor mobile connections from the remote end. While mobile carrier networks continue to improve, we recommend following up with the remote end promptly to see if they may have been in a questionable signal zone.

If the source of the disconnect is not clearly identifiable, working with the SharpenCare team can help us get closer. Our logging can confirm whether the dropped call was initiated by the agent side, system side, or remote end. If you need to track down an important dropped call and need some assistance, please raise a request.

Applicable guides

Delayed audio

Potential Causes

Common Solutions

  • Carrier service latency

  • Agent ISP service latency

  • Agent local network latency

  • Agent browser degradation (memory exhaustion)

  • Agent workstation degradation (CPU/memory exhaustion)

  • Sharpen voice infrastructure degradation

  • Call the remote party back. Is the experience consistent, and does it happen when called from a non-Sharpen phone? If so, it can be concluded that the delay is on the remote end.

  • Open a request with the agent or organization ISP(Internet Service Provider). While ISPs can sometimes be slow to reveal information to confirm interruption, the best ammunition we have is timely reports. If something has gone beyond 24 hours from the impact, we’re not likely to get the information we need. Make sure to be specific with impact and ask very clearly whether any interruptions occurred.

  • Open a request with the local network team. Power exists in numbers. If this impacted many agents on the same network, there’s a higher likelihood the issue exists within the local network.

  • If suspecting workstation degradation, leverage the agent’s operating system resource monitor to validate sufficient memory, CPU, and disk exists. While there’s no specific rule here, if any of these resources is encroaching on 80% utilization, it would be best to increase availability.

  • Work with SharpenCare to observe and troubleshoot the health of webRTC and Voice Processor infrastructure. We may ask to retrieve browser logs (if available) to more specifically identify the issue.

...