How to interpret the "TCP DELAY" report for event diagnosis in wolkvox?
Learn effective network monitoring techniques, improving the security and performance of your technology infrastructure.
Written by Sandra Ocaña
Updated at August 5th, 2026
Table of Contents
Symptom or Need
As a supervisor or administrator of the operation, you need to interpret the metrics presented in the TCP DELAY network report to understand the meaning of each column, identify the allowable operability thresholds, and determine how deviations in these values affect call quality (interruptions, robotic audio, or disconnections).
Context / Scenarios
Wolkvox is a cloud-based solution (CCaaS) whose optimal performance depends on the stability of your local network as a client and your internet connection. To facilitate real-time technical validation, Wolkvox provides the TCP Delay report, which evaluates the performance between the agent's workstation and the server.
Access to the report
Supervisors can log in using the following URL:
https://wv{serverXXXX}.wolkvox.com/ipdialbox/latencia.php?id={operación}
Upon logging in, you will be able to view a real-time metrics table for each agent.
Types of traffic in monitoring
It is essential to differentiate data flows to identify whether an event is general to the network or specific to a user (e.g., Wi-Fi connection or without QoS prioritization):
- TX (Transmission): shows how the server transmits packets (impacts what the agent/client hears).
- RX (Reception): shows how the server receives the packets (impacts what the server processes or listens to).
Answer / Solution
When accessing the TOTAL TCP DELAY BY AGENT report, you should analyze the following key fields:
1. USER and STATUS
What does it show?
- USER: Indicates the agent's telephone extension (e.g., 3971 or 3995).
- STATUS: reflects the registration status of the extension with the server (e.g., disabled if it is disabled or disconnected).
What is it for?
It allows you to quickly and individually locate which agent is experiencing network anomalies and verify if its main connection to the server is still active.
2. TX JITTER and RX JITTER (Packet Time Variation)
What does it show?
- The variation in the arrival time of data packets.
- TX JITTER : measures the traffic transmitted by the server (affects what the client hears).
- RX JITTER: measures the traffic received by the server (affects what it processes).
Allowed threshold : Must be < 30 ms.
What could it affect?
Real-time audio is highly sensitive to jitter. Exceeding 30 ms results in severe degradation: robotic, out-of-sync, or choppy audio.
3. TX PLOSS and RX PLOSS (Packet Loss)
What does it show?
- Percentage of data packets that are lost during the Transmission (TX) or Reception (RX) path and do not reach the server.
Permissible threshold : Must be < 1%.
What could it affect?
This results in silences or interruptions. If the loss exceeds 1%, syllables, words, or entire phrases will be lost during communication with the client.
4. Global and Complementary Metrics to be taken into account
- TCP Delay (Latency):
Description: This is the total time for acknowledgment and transmission.
Allowed threshold: Must be strictly < 500 ms, although the ideal connectivity scenario is < 150 ms or 100 ms.
Impact on operations: Voice delay (interlocutors speak at the same time), call drops due to timeout, and HTTP/WebSocket synchronization failures.
- Network Rejection (NR):
Description: This corresponds to the number of calls not delivered to the agent due to network fluctuations.
Permissible threshold: Must be < 1%.
Impact on operations: Increase in missed calls, unassigned interactions, and impact on productivity.
What can TCP Delay affect?
- Interface and Agent Script failures: Latency prevents the timely synchronization of HTTP and WebSocket requests. As a result, the interface and heavy dynamic elements (such as the Agent Script) fail to render correctly on screen or capture call data, negatively impacting the user experience.
- Voice Degradation: Low audio quality, micro-cuts or silent calls.
- False States (Interface Lag) : The agent sees on its screen that it is "Ready", but due to lack of synchronization, on the supervisor's monitor the system has logged it out due to inactivity (Logout).
- Interaction Interruptions: Unexpected drops in calls or chat sessions.
- Increased Network Rejection.
Immediate Actions for Alerts in the TCP Delay Report: If you detect metrics outside the allowed thresholds, validate the following points with the IT team and/or the agent:
- Physical connection: Ensures the use of a wired network (Ethernet) and avoids Wi-Fi or mobile data connections.
- Quality of Service (QoS): Verifies the implementation of QoS policies on the local network, prioritizing voice traffic with DSCP EF dialing (46).
- Bandwidth: Ensure the required capacity:
Corporate LAN environments: 400 Kbps synchronous per agent.
Working from home (Teleworking): Minimum 30 Mbps dedicated.
- ISP Stability: Confirms that there is no packet loss on the Internet Service Provider (ISP) links to the wolkvox servers.
- Network Profile: Check in the Event Viewer that the network profile does not show any disconnections or recurring status changes.
Recommendation
It is essential that the IT/Systems department continuously monitor network performance through latency reporting. This allows them to identify and mitigate bottlenecks before they impact operations and customer service levels.