This topic lists major metrics for Server Load Balancer (SLB) Layer 4 and Layer 7 protocols and provides corresponding descriptions. You can set the thresholds of metrics, and create alert rules to trigger alerts.

Precautions

  • Call the Cloud Monitor API to query monitoring data from the last 31 days.
  • Set Project to acs_slb_dashboard, and set the data sampling period to 60 seconds. The Period parameter can be set only to a multiple of 60.
  • Set the Dimensions parameter to a JSON string. Example: [{"instanceId":"lb-bp1r92vzpemy099f******"}].

    In Dimensions, instanceId specifies the ID of the SLB instance, port specifies the port number of the SLB instance, and vip specifies the service address of the SLB instance.

Layer 4 protocol metrics

Metric Description Unit Dimension Statistics
HeathyServerCount The number of healthy ECS instances. Count instanceId Average, Minimum, and Maximum
UnhealthyServerCount The number of unhealthy ECS instances. Count instanceId Average, Minimum, and Maximum
PacketTX The number of outbound packets sent per second on a port. Count/Second instanceId, port, and vip Average, Minimum, and Maximum
PacketRX The number of inbound packets received per second on a port. Count/Second instanceId, port, and vip Average, Minimum, and Maximum
TrafficRXNew The amount of data received per second on a port. bit/s instanceId, port, and vip Average, Minimum, and Maximum
TrafficTXNew The amount of data sent per second on a port. bit/s instanceId, port, and vip Average, Minimum, and Maximum
ActiveConnection The number of active connections to an SLB instance on a port, which indicates the number of connections that clients set up to access the SLB instance.

It indicates the number of TCP connections that are in the ESTABLISHED state. If persistent connections are used, a connection can transfer multiple file requests simultaneously.

Count instanceId, port, and vip Average, Minimum, and Maximum
InactiveConnection The number of inactive connections to an SLB instance on a port, which indicates the number of idle connections to the SLB instance.

It indicates the number of TCP connections that are not in the ESTABLISHED state. You can run the netstat -an command to view the connections for both Windows and Linux instances.

Count instanceId, port, and vip Average, Minimum, and Maximum
NewConnection The number of new TCP connections established between clients and an SLB instance within a statistical period. Count instanceId, port, and vip Average, Minimum, and Maximum
MaxConnection The number of established TCP connections. Count instanceId, port, and vip Average, Minimum, and Maximum
DropConnection The number of connections dropped per second. Count/Second instanceId, port, and vip Average, Minimum, and Maximum
DropPacketRX The number of inbound packets dropped per second. Count/Second instanceId, port, and vip Average, Minimum, and Maximum
DropPacketTX The number of outbound packets dropped per second. Count/Second instanceId, port, and vip Average, Minimum, and Maximum
DropTrafficRX The amount of inbound traffic dropped per second. bit/s instanceId, port, and vip Average, Minimum, and Maximum
DropTrafficTX The amount of outbound traffic dropped per second. bit/s instanceId, port, and vip Average, Minimum, and Maximum
InstanceActiveConnection The number of active connections per second on an instance. Count/Second instanceId Average, Minimum, and Maximum
InstanceDropConnection The number of connections dropped per second on an instance. Count/Second instanceId Average, Minimum, and Maximum
InstanceDropPacketRX The number of inbound packets dropped per second on an instance. Count/Second instanceId Average, Minimum, and Maximum
InstanceDropPacketTX The number of outbound packets dropped per second on an instance. Count/Second instanceId Average, Minimum, and Maximum
InstanceDropTrafficRX The amount of inbound traffic dropped per second on an instance. bit/s instanceId Average, Minimum, and Maximum
InstanceDropTrafficTX The amount of outbound traffic dropped per second on an instance. bit/s instanceId Average, Minimum, and Maximum
InstanceInactiveConnection The number of idle connections per second on an instance. Count/Second instanceId Average, Minimum, and Maximum
InstanceMaxConnection The maximum number of concurrent connections per second on an instance. Count/Second instanceId Average, Minimum, and Maximum
InstanceNewConnection The number of new connections per second on an instance. Count/Second instanceId Average, Minimum, and Maximum
InstancePacketRX The number of request packets received by an SLB instance per second. Count/Second instanceId Average, Minimum, and Maximum
InstancePacketTX The number of packets sent by an SLB instance per second. Count/Second instanceId Average, Minimum, and Maximum
InstanceTrafficRX The traffic consumed to access an SLB instance from the Internet. bit/s instanceId Average, Minimum, and Maximum
InstanceTrafficTX The traffic consumed by an SLB instance to access the Internet. bit/s instanceId Average, Minimum, and Maximum
InstanceMaxConnectionUtilization The maximum utilization of connections. % instanceId Average, Minimum, and Maximum
InstanceNewConnectionUtilization The utilization of new connections. % instanceId Average, Minimum, and Maximum

Layer 7 protocol metrics

Metric Description Unit Dimensions Statistics
Qps The queries per second (QPS) through a port, which indicates the number of HTTP and HTTPS requests processed per second. Count/Second instanceId, port, and vip Average
Rt The average latency of requests on a port, which indicates the average response time of an SLB instance. ms instanceId, port, and vip Average
StatusCode2xx The number of 2xx status codes returned by an SLB instance to the client through a port. Count/Second instanceId, port, and vip Average
StatusCode3xx The number of 3xx status codes returned by an SLB instance to the client through a port. Count/Second instanceId, port, and vip Average
StatusCode4xx The number of 4xx status codes returned by an SLB instance to the client through a port. Count/Second instanceId, port, and vip Average
StatusCode5xx The number of 5xx status codes returned by an SLB instance to the client through a port. Count/Second instanceId, port, and vip Average
StatusCodeOther The number of other status codes returned by an SLB instance to the client through a port. Count/Second instanceId, port, and vip Average
UpstreamCode4xx The number of 4xx status codes returned by backend servers to an SLB instance through a port. Count/Second instanceId, port, and vip Average
UpstreamCode5xx The number of 5xx status codes returned by backend servers to the client through a port. Count/Second instanceId, port, and vip Average
UpstreamRt The average latency of requests sent by backend servers to the proxy through a port. ms instanceId, port, and vip Average
InstanceQps The queries per second (QPS) on an instance, which indicates the number of HTTP and HTTPS requests processed per second. Count/Second instanceId Average
InstanceRt The average response latency of an instance. ms instanceId Average
InstanceStatusCode2xx The number of 2xx status codes returned by an SLB instance to the client. Count/Second instanceId Average
InstanceStatusCode3xx The number of 3xx status codes returned by an SLB instance to the client. Count/Second instanceId Average
InstanceStatusCode4xx The number of 4xx status codes returned by an SLB instance to the client. Count/Second instanceId Average
InstanceStatusCode5xx The number of 5xx status codes returned by an SLB instance to the client. Count/Second instanceId Average
InstanceStatusCodeOther The number of other status codes returned by an SLB instance to the client. Count/Second instanceId Average
InstanceUpstreamCode4xx The number of 4xx status codes returned by backend servers to an SLB instance. Count/Second instanceId Average
InstanceUpstreamCode5xx The number of 5xx status codes returned by backend servers to an SLB instance. Count/Second instanceId Average
InstanceUpstreamRt The average latency of requests sent from backend servers of an SLB instance to a proxy. ms instanceId Average
InstanceQpsUtilization The QPS utilization. % instanceId Average, Minimum, and Maximum