1. Hardware Monitor

In Data Collection, we introduced the common data sources. As a monitor framework, Open-Falcon can collect the monitor index data of any system as long as they are converted to the standard format of Open-Falcon.

The data collection of hareware can be done through HWCheck.

2. HWCheck

Falcon-Agent needs to be installed, then modules like Rvadmin Tool can perform hareware monitor.

It only supports dell physical machines. The following metrics can be monitored:

cpu memory RAID physical disk virtual disk RAID battery BIOS mainboard battery fan voltage mainboard temperature cpu temperature

3. Installation Instruction

  1. Configure the official repo of Dell and install dependencies like srvadmin and so on
#Reference: http://linux.dell.com/repo/hardware/latest/
wget -q -O - http://linux.dell.com/repo/hardware/latest/bootstrap.cgi | bash

yum install srvadmin-omacore srvadmin-omcommon srvadmin-storage-cli smbios-utils-bin lm_sensors dmidecode cronie
# Enable srvadmin service
/opt/dell/srvadmin/sbin/srvadmin-services.sh enable
/opt/dell/srvadmin/sbin/srvadmin-services.sh restart
# Configure lm-sensors
echo yes | /usr/sbin/sensors-detect

3.1. You can also pack the rpm to simplify the deployment

git clone https://github.com/51web/hwcheck hwcheck-0.2
tar czf hwcheck-0.2.tar.gz hwcheck-0.2
rpmbuild -tb hwcheck-0.2.tar.gz

4. How to Use

4.1. Parameter Information

Execute hwcheck without parameters will directly print the detailed monitor data.

hwcheck -d      # print the metrics information,which means the data pushed to falcon-agent
        -p      # push data falcon-agent
        -s      # the value of STEP in data pushing, which means frequency; the default value is 600 seconds
        -m      # one specified metric

4.2. Configure Crontab

Configure cron to test regularly, like:

cat /etc/cron.d/hwcheck

18 * * * * root /usr/bin/hwcheck -s 3600 -p >/dev/null 2>&1 &

The value of STEP is 3600, which means run a test each hour.

4.3. Configure Alarm Strategy om Falcon-Portal

The metrics pushed by hwcheck to falcon-agent all begin with hw, like hw.cpu_temp, except for the acutal number of the temperature.

In other value of metric: 0 means malfunction,1 means warning,2 meas OK. For example, configure the following strategies in Portal:

metric/tags/note condition max P
hw.bios [C1E in BIOS中/Cstate enabled] all(#2)<2 1 4
hw.board_temp [mainboard overheating] all(#3)>=35 1 4
hw.cmos_bat [mainboard bettery issue] all(#3)<2 1 4
hw.cpu [CPU possible breakdown] all(#2)==1 1 4
hw.cpu [Impoetant: CPU severe breakdown] all(#2)==0 2 0
hw.fan [fan breakdown] all(#3)<2 1 4
hw.memory [memory possible breakdown] all(#1)==1 1 4
hw.memory [Important: memory severe breakdown] all(#1)==0 2 0
hw.pdisk [Important: physical disk severe breakdown] all(#1)==0 2 0
hw.raidcard [RAID warning] all(#2)==1 1 4
hw.raidcard [Important: RAID severe breakdown] all(#1)==0 2 0
hw.raidcard_bat [RAID bettery warning] all(#2)==1 1 4
hw.raidcard_bat [Important: RAID battery severe breakdown] all(#2)==0 2 0
hw.vdisk [disk array warning] all(#2)==1 1 4
hw.vdisk [Important: disk array severe breakdown] all(#2)==0 2 0
Copyright 2015 - 2018 Xiaomi Inc. all right reserved,powered by Gitbook该文件修订时间: 2018-06-25 10:30:02