Docs

Elastic Compute Service

Monitoring metrics

Viewing TCP connection monitoring data

To allow you to view the TCP connection monitoring data for the instance operating system, Volcano Engine provides you with the TCP connection monitoring feature. This allows you to view metrics of the TCP connection state.

View TCP connection monitoring data

Usage notes

Before using TCP connection monitoring, you need to take the following actions:

Currently, you can only view TCP connection monitoring information for instances with certain operating systems. Please make sure that the operating system of your instance has one of the following versions:

Type	Released version
CentOS	CentOS Stream 9 64 bit CentOS Stream 8 64 bit CentOS 8.x 64 bit CentOS 7.2 64 bit and later versions
Ubuntu	Ubuntu Server 22.04 LTS 64 bit Ubuntu Server 20.04 LTS 64 bit Ubuntu Server 16.04 LTS 64 bit
Debian	Debian 11 64 bit 64-bit Debian 9
Fedora	Fedora 37 64 bit
veLinux	veLinux 1.0 64 bit quick start image veLinux 1.0 CentOS compatible version with 5.10 kernel 64 bit veLinux 1.0 64 bit Hardened Image for Security Protection veLinux 1.0 CentOS compatible version 64 bit veLinux 1.0 with 5.10 kernel 64 bit veLinux 1.0 64 bit
AlmaLinux	AlmaLinux 9.1 64 bit AlmaLinux 8.7 64 bit

Please check whether the instance has been installed with the Cloud Monitor plugin according to the ECS monitoring guide. If not, please refer to Installing plugin to install it.

You need to manually enable the TCP connection monitoring feature. For details, refer to TCP connection metrics missing.

The collection frequency for TCP connection monitoring metrics is 30s. This means that TCP connection monitoring metrics are collected every 30s.

Procedure

In the top navigation bar, select the project and region of the target instance.

In the left-side navigation pane, select "Instances & Images > Instances".

Click the name of the target instance to go to the instance details page.

Select the "Monitoring > TCP connection monitor" tab. On this tab, you can view metrics such as TCP bytes sent/received and connection retransmission counts for connections between local and remote IP addresses. For more metric details, refer to Viewing TCP connection monitoring metrics.

warning

After the instance starts, wait approximately 10 minutes and then view the relevant metric data within the instance.

Related operation

You can also create an alarm strategy for TCP connection monitoring. This helps you to learn whether processes on your ECS instance are alive in a timely manner and thereby ensures processes run properly.

View TCP connection monitoring metrics

tip

The collection frequency for TCP connection monitoring is 30 seconds. This means that relevant TCP connection monitoring data is collected every 30 seconds.

Metric	Explanation	Unit
ProcTxBytes	Network bytes sent by processes	Byte
ProcRxBytes	Network bytes received by processes	Byte
ProcTxPackets	Network packets sent by processes	Count
ProcRxPackets	Network packets received by processes	Count
ConnTxBytes	Network bytes sent by TCP connections	Byte
ConnRxBytes	Network bytes received by TCP connections	Byte
ConnTxPackets	Network packets sent by TCP connections	Count
ConnRxPackets	Network packets received by TCP connections	Count
ConnRetransmits	Retransmission count of TCP connections	Count
ConnRTT	RTT of TCP connections	μs
ConnRTTVar	RTT variance of TCP connections	μs

Last updated: 2026.06.18 17:49:51