业务报警订单提交异常，页面一直没有反应，排查后是事务没有提交或者回滚导致，想到如果及时监控事务的运行状态报警出来，那么就
序言：
业务报警订单提交异常，页面一直没有反应，排查后是事务没有提交或者回滚导致，想到如果及时监控事务的运行状态报警出来，那么就可以及时排查出问题所在，方便运营处理，所以自己就弄了一个shell脚本放在nagios来处理事务报警情况。
1,编写事务监控脚本
#!/bin/bash
# author: tim.man
# version: 1.0
# desc: check the running transaction over
st_ok=0 
st_wr=1 
st_cr=2 
st_uk=3
time_trx=10
# 提示信息
print_help() { 
    echo $progname -w int -c int 
    echo options: 
    echo   -w/--warning) 
    echo     sets a warning number 
    echo   -c/--critical) 
    echo     sets a critical level for io 
    exit $st_uk 
}
while test -n $1; do 
    case $1 in 
        -help|-h) 
            print_help 
            exit $st_uk 
            ;; 
        --warning|-w) 
            warning=$2 
        shift 
            ;; 
        --critical|-c) 
            critical=$2 
        shift 
            ;; 
        *) 
            echo unknown argument: $1 
            print_help 
            exit $st_uk 
            ;; 
        esac 
    shift 
done
get_wcdiff() { 
    if [ ! -z $warning -a ! -z $critical ] 
    then 
        wclvls=1
if [ ${warning} -gt ${critical} ] 
        then 
            wcdiff=1 
        fi 
    elif [ ! -z $warning -a -z $critical ] 
    then 
        wcdiff=2 
    elif [ -z $warning -a ! -z $critical ] 
    then 
        wcdiff=3 
    fi 
}
# 脚本判断
val_wcdiff() { 
    if [ $wcdiff = 1 ] 
    then 
        echo please adjust your warning/critical thresholds. the warning must be lower than the critical level! 
        exit $st_uk 
    elif [ $wcdiff = 2 ] 
    then 
        echo please also set a critical value when you want to use warning/critical thresholds! 
        exit $st_uk 
    elif [ $wcdiff = 3 ] 
    then 
        echo please also set a warning value when you want to use warning/critical thresholds! 
        exit $st_uk 
    fi 
}
get_wcdiff 
val_wcdiff
# 统计mysql的事务中最大运行时间
max_over_time=`/usr/local/mysql/bin/mysql --user=nagios --password=nagiosq@xxx  -ns /usr/local/mysql/mysql.sock -e select time_to_sec(timediff(now(),t.trx_started)) from information_schem
a.innodb_trx t where time_to_sec(timediff(now(),t.trx_started))>$time_trx order by time_to_sec(timediff(now(),t.trx_started)) desc limit 1; |awk '{print $1}'`
# 如果当前没有running的事务，则直接赋值为0，以免下面if判断出错
if [ ! -n $max_over_time ];then max_over_time=0
fi
# 取得当前所以阻塞的事务数量
num_trx=`/usr/local/mysql/bin/mysql --user=nagios --password=nagiosq@xxx  -ns /usr/local/mysql/mysql.sock -e select count(1) from information_schema.innodb_trx t where time_to_sec(timedif
f(now(),t.trx_started))>$time_trx; |awk '{print $1}'`
if [ -n $warning -a -n $critical ] 
then 
    if [ `expr $max_over_time \> $warning` -eq 1 -a `expr  $max_over_time \    then 
        echo warning - $num_trx transactions running,go over for $max_over_time seconds 
        exit $st_wr 
    elif [ `expr $max_over_time \> $critical` -eq 1 ] 
    then 
        echo critical- $num_trx transactions runnning,go over for $max_over_time seconds 
        exit $st_cr 
    else 
        echo ok- transactions ran successfully. 
        exit $st_ok 
    fi 
fi