Grafana influxdb simple script

InfluxDBtools-Linux Document resources -CSDN download One for connecting InfluxDB Simple tools for , Decompression is used immediately. , password 666666 password 666666 More download resources 、 For learning materials, please visit CSDN Download channel .https://download.csdn.net/download/zhangxueleishamo/84196848

One 、 demand
machine IP：ip1     ip2   ip3   
File storage path ：/data/logs/bireport/statistics  
File update time ： Every morning 1 spot
The contents of the document ： The first line is the number of timeouts , The second line is the total number of calls

#rsync client
1 2 * * * sh /root/rsync_bi.sh
cat /root/rsync_bi.sh
#!/bin/bash
passwdfile="/root/.rsync.passwd"
server="[email protected]::backup"
server_patch=$server/ip1
pathfile=/data/logs/bireport/statistics 
/usr/bin/rsync -qazu --password-file=$passwdfile $pathfile $server_patch/

#grafana server 
30 2 * * * sh /data/bireport/bi_request.sh
[root@localhost bireport]# cat bi_request.sh 
#!/bin/bash
pathdir="/data/bireport"
filename="bi_history"

ip_num=(ip1
ip2
ip3)

datetime1=`date -d "yesterday" +%Y%m%d`
datetime=`date -d "$datetime1 0 +8 hour" +"%Y%m%d %H"`
timeStamp_y=`date -d "$datetime" +%s`
#hour_8=28800000
timeStamp=`expr $timeStamp_y \* 1000`

for ip in ${ip_num[@]} 
do
	if [ ! -f "$pathdir/$ip/statistics" ]; then
  		echo "ERROR: $pathdir/$ip/statistics is not exist!"
  		exit 1
	else
		timeout_num=`cat $pathdir/$ip/statistics |awk 'NR<2'`
		request_num=`cat $pathdir/$ip/statistics |awk 'NR>1'`
		rate_num=`echo  "scale=4; a = $timeout_num / $request_num; if (length(a)==scale(a)) print 0;print a"  | bc`
		echo $timeStamp $ip $request_num $timeout_num $rate_num >> $pathdir/$filename
		/usr/bin/curl -i -XPOST 'http://127.0.0.1:8086/write?db=bi_request_info&precision=ms' -u admin:cAauFWEtLerqH4EP --data-binary 'bi_info,host_id='$ip' request_num='$request_num',timeout_num='$timeout_num',rate_mun='$rate_num' '$timeStamp''
		
		mv $pathdir/$ip/statistics  $pathdir/$ip/$timeStamp"statistics"
		
	fi
done

notes ： Cause right influxdb I'm not familiar with the time format , Waste a lot of time ;

Two 、influxdb In time format

influxdb Three time formats are supported ,epoch_time Outside , And support rfc3339_date_time_string and rfc3339_like_date_time_string.

1、epoch_time Format
A time stamp indicates , We usually use 10 Bit and 13 position , stay influxdb The timestamp used in is 19 position , The unit is ns( nanosecond )

2、rfc3339_date_time_string Format
This format is ：'YYYY-MM-DDTHH:MM:SS.nnnnnnnnnZ', among nnnnnnnnn It's optional , If you don't write, it will be set to 000000000. Be careful , If you use this time format , You need to use single quotes ’ Enclose time .

3、rfc3339_like_date_time_string
This format is ：'YYYY-MM-DD HH:MM:SS.nnnnnnnnn' , among HH::MM:SS.nnnnnnnnn It can be omitted , Must be enclosed in single quotes .

4、 Query by timestamp
Second level ：
select * from disk where time >= 1542954639s and time <= 1542964713s
millisecond :
select * from disk where time >= 1542954639000ms and time <= 1542964714000ms
Nanosecond ：
select * from disk where time >= 1542954639000000000ms and time <= 1542964714000000000ms

5、 Adjust time zone query - Beijing time.
select * from disk where time >= '2018-11-23 14:30:39' and time <= '2018-11-23 14:32:32' tz('Asia/Shanghai')

notes ：
1、 because grafana To use UTC Time , So... Is added in this example 8 Hours
2、curl When inserting data , Such as insertion time The time format needs to be set &precision=ms For milliseconds , Or for s
3、 Inserted data --data-binary You need to add single quotation marks after ''