LogCollector v2.4.1 and later versions support collecting host monitoring data to the metrics topic according to the collection configuration. This topic describes the steps for collecting host monitoring.

Prerequisites

An IAM user is created and granted permissions related to data collection, such as RemoteWrite. For details about permissions, see Data collection.

A project and metric topic have been created. For details, see Metric topic.

A host group is created and the server heartbeat status is in normal state. For detailed steps, see Create host group (IP address) or Create host group (Host identification).

The server where LogCollector is installed must be able to access port 80 and port 443 on the remote server.

LogCollector version 2.4.1 or later version has been installed on the server. For detailed steps, see Install LogCollector (host machine).

Steps

Step 1 Select a log space

Log in to TLS console.

Select the region where log service is located from the top navigation bar.

On the left navigation bar, select Common functions > Log collection.

On the log collection page, click Metrics collection, and then click Host monitoring.

Select the project and metrics topic, and then click Next step: Select host group.

The collected metric data will be stored in this metric topic.

Step 2 Select host group

Select the host group you need to collect metrics from in the All host group section.

If no host group is available in the list, you need to create a host group first. For detailed instructions, see Create host group (Host identification) and Create host group (IP address).

After confirming the host group in the Selected host group section, click Next Step: Collection rule.

Step 3 Configure collection rule

Enter the Rule name.

Length must be between 3 and 64 characters.

Only Chinese characters, letters, numbers, underscores (_), and hyphens (-) are allowed.

Can not start or end with a hyphen (-).

Enter collection configuration.

Enter the monitoring configuration information for the collection host. Example:

Default settings

{
"input": "host_metric",
"detail": {
"scrape_interval": 30000
}
}

Configure metadata collection for ECS hosts

If deployed on ECS, you can obtain host metadata. For details, see view instance metadata. Some parameter examples are shown below:

{
"input": "host_metric",
"detail": {
"scrape_interval": 30000
},
"processors": [
{
"processor_appender": {
"sort_labels": true,
"value": "|cloud_instance_id#$#{{instance_id}}|cloud_instance_name#$#{{$__cloud_instance_name__}}|cloud_region_id#$#{{region_id}}|cloud_zone_id#$#{{availability_zone}}"
}
}
]
}

Parameters	Example value	Notice
cloud_instance_id	i-yeji7a641ss6ipokxxxx	Cloud server instance ID.
cloud_instance_name	ECS-test	Name of the cloud server instance.
cloud_region_id	cn-xxx	Region ID where the cloud server instance is located.
cloud_zone_id	cn-xxx-b	Availability zone ID where the cloud server instance is located.

Click Submit.

Next steps

After you create a collection configuration, LogCollector collects metric data and saves it to the specified metric topic. You can perform query and analysis in the metric topic. For details, see Query and analyze metric topic.

Description

The metrics collected for host CPU, memory, load, disk, and network are explained below.

Host CPU metrics

Metric name	Description
cpu_count	Number of CPU cores.
cpu_util	CPU utilization percentages. Idle, wait, and steal are excluded.
cpu_guest_util	Percentage of guest time.
cpu_guestnice_util	Percentage of nice guest time.
cpu_irq_util	Percentage of hard IRQ processing time.
cpu_nice_util	Percentage of nice time.
cpu_softirq_util	Percentage of soft IRQ processing time.
cpu_steal_util	Percentage of CPU waiting time (Steal time) on the host machine.
cpu_sys_util	Percentage of kernel mode (System time).
cpu_user_util	Percentage of user mode (User time).
cpu_wait_util	Percentage of waiting for IO (Waiting time).

Memory metrics

Metric name	Description
mem_util	Memory usage.
mem_cache	Allocated but unused memory (bytes).
mem_free	Unused memory (bytes).
mem_available	Available memory (bytes).
mem_used	Used memory (bytes).
mem_swap_util	Swap memory utilization.
mem_total	Total memory (byte).

Disk metrics

Metric name	Description
disk_rbps	Hard disk read throughput per second (byte/s).
disk_wbps	Hard disk write throughput per second (byte/s).
disk_riops	Number of disk reads per second (ops/s).
disk_wiops	Number of disk writes per second (ops/s).
disk_rlatency	Average read latency (ms).
disk_wlatency	Average write latency (ms).
disk_util	IO utilization.
disk_space_usage	Disk usage percentage.
disk_inode_usage	Inode utilization.
disk_space_used	Disk used capacity (byte/s).
disk_space_total	Total disk capacity (byte/s).
disk_inode_total	Total number of inodes.
disk_inode_used	Number of used inodes.

NET metrics

Metric name	Description
net_drop_util	Ratio of dropped packets to total packets
net_err_util	Ratio of erroneous packets to total packets
net_in	Network receiving rate (byte/s).
net_in_pkt	Packets received per second (count/s).
net_out	Network transmit rate (byte/s).
net_out_pkt	Packets sent per second (count/s).

TCP metrics

Metric name	Description
protocol_tcp_established	Established connections (count).
protocol_tcp_insegs	Total number of packets received.
protocol_tcp_outsegs	Number of packets sent.
protocol_tcp_retran_segs	Number of retransmitted packets.
protocol_tcp_retran_util	Ratio of retransmitted packets to total packets sent.

System metrics

Metric name	Description
system_boot_time	System boot time (s).
system_load1	System average load, 1-minute load average.
system_load5	System average load, 5-minute load average.
system_load15	System average load, 15-minute load average.

Torch Log Service