Metricbeat deployment and monitoring of Linux system metrics summary

Metricbeat Light-weight indicator collector

Used to collect metrics from systems and services. From CPU to memory, from Redis to Nginx,metricbeat, you can deliver a variety of system and service statistics in a light-weight way.

System-level monitoring, more concise

Deploy Metricbeat to all your Linux, Windows, and MAC hosts and connect it to Elasticsearch: you get system-level CPU usage, memory, file system, disk IO and network IO statistics, and get the same Statistics for each process similar to the top command on the system. Explore [Online Demo] (https://demo.elastic.co/app/kibana#/dashboard/Metricbeat-system-overview?_g= ()

Installing Metricbeat

wget https://artifacts.elastic.co/downloads/beats/metricbeat/metricbeat-6.3.2-linux-x86_64.tar.gz
tar -zxvf metricbeat-6.3.2-linux-x86_64.tar.gz
mv metricbeat-6.3.2-linux-x86_64 metricbeat

Configure Metricbeat

metricbeat.config.modules:
   path: $ {path.config} /modules.d / *. yml
   reload.enabled: false
setup.template.settings:
   index.number_of_shards: 1
   index.codec: best_compression
setup.kibana:
   host: "localhost: 5602"
output.elasticsearch:
   hosts: ["192.168.1.1:29200"]
## Enable sytem monitoring
$ cat modules.d / system.yml
-module: system
   period: 10s
   metricsets:
     -cpu
     -load
     -memory
     -network
     -process
     -process_summary
     #-core
     -diskio
     -socket
   processes: [‘. *‘]
   process.include_top_n:
     by_cpu: 5 # include top 5 processes by CPU
     by_memory: 5 # include top 5 processes by memory

-module: system
   period: 1m
   metricsets:
     -filesystem
     #-fsstat
   processors:
   -drop_event.when.regexp:
       system.filesystem.mount_point: ‘^ / (sys | cgroup | proc | dev | etc | host | lib) ($ | /) ′

-module: system
   period: 1m
   metricsets:
     -uptime 

Start Kibana, ES, metricbeat

/usr/local/metricbeat/metricbeat -e -c /usr/local/metricbeat/metricbeat.yml

Collect summaries using Granfan visual system metrics

Indicator Type	Indicators	indicator meaning
Cpu	system.cpu.total.pct	Total CPU Usage Percentage
Cpu	System.cpu.cores	Number of CPU Cores
Cpu	system.cpu.iowait.pct	Percentage of CPU time waiting for input and output
Cpu	system.cpu.user.pct	Percentage of CPU occupied by user space
Cpu	system.cpu.system.pct	Percentage of CPU consumed by kernel space
Cpu	system.cpu.nice.pct	Process changes consume CPU percentage
Cpu	system.cpu.idle.pct	Percentage of idle CPU
Memory	System.memory.used.bytes	Memory usage Size
Memory	system.memory.used.pct	Memory usage percent
Memory	System.memory.free	Memory remaining size
Memory	System.memory.total	Total Memory Size
Memory	system.memory.swap.used.pct	Swap memory usage percent
Memory	System.memory.swap.used.bytes	Swap memory usage Size
Memory	System.memory.swap.free	Swap remaining memory
Memory	System.memory.swap.total	Total Swap Memory size
Network	System.network.name	Nic Name
Network	System.network.in.packets	Number of network card ingress packages
Network	System.network.in.errors	NIC Ingress error Packet number
Network	system.network.in.dropped	Network card ingress reject packet number
Network	System.network.in.bytes	Network Card Entry Package size
Network	System.network.out.packets	Number of network card egress NIC Packets
Network	System.network.out.bytes	Network card egress Packet size
Network	System.network.out.errors	Network card egress Error packet number
Network	system.network.out.dropped	Number of rejection packets for network card egress
Load	System.load.1	1-minute system average load
Load	System.load.5	5-minute system average load
Load	system.load.15	15-minute system average load
Process_summary	system.process.summary.stopped	Stop process
Process_summary	System.process.summary.zombie	Zombie Process
Process_summary	System.process.summary.unknown	Stateless process
Process_summary	System.process.summary.total	Total number of processes
Process_summary	System.process.summary.sleeping	Hibernation process
Process_summary	System.process.summary.running	Running processes
Uptime	system.uptime.duration.ms	System Run time
Socket	System.socket.local.ip	Native IP
Diskio	System.diskio.iostat.read.per_sec.bytes	Amount of data read from the device (drive expressed) per second (kb_read/s)
Diskio	System.diskio.iostat.write.per_sec.bytes	Amount of data written to the device (drive expressed) per second (KB_WRTN/S)
Diskio	System.diskio.iostat.read.request.per_sec	Number of sectors read per second (RSEC/S)
Diskio	System.diskio.iostat.write.request.per_sec	Number of sectors written per second (wsec/s)
Diskio	System.diskio.iostat.read.request.merges_per_sec	How much of this device-dependent read request is merge (rrqm/s) per second
Diskio	System.diskio.iostat.write.request.merges_per_sec	How much of this device-dependent write request is merge (wrqm/s) per second
Diskio	System.diskio.iostat.await	Average time (in microseconds) of processing per IO request
Diskio	System.diskio.read.bytes	Total amount of data read (Kb_read)
Diskio	System.diskio.write.bytes	Total amount of data written (KB_WRTN)
FileSystem	System.filesystem.device_name	File System device Name
FileSystem	System.filesystem.free	Disk space remaining
FileSystem	System.filesystem.mount_point	Disk Mount Partition
FileSystem	System.filesystem.total	Total disk size
FileSystem	system.filesystem.used.pct	Disk usage
FileSystem	System.filesystem.used.bytes	Disk usage Size
FileSystem	System.filesystem.used.bytes	Disk usage Size

Drawing templates

System-metrics

Effects such as

Reference:

• Https://www.elastic.co/cn/products/beats/metricbeat
• Https://www.elastic.co/guide/en/beats/metricbeat/current/exported-fields-system.html
• Personal blog

Metricbeat deployment and monitoring of Linux system metrics summary