简体   繁体   English

如何调试收集到的执行脚本

[英]How to debug exec script for collectd

I wrote a little script to collect the power state (standby, actice) of my disk. 我写了一个小脚本来收集磁盘的电源状态(备用,actice)。 It is based on this script: https://github.com/collectd/collectd/blob/master/contrib/exec-smartctl 它基于以下脚本: https : //github.com/collectd/collectd/blob/master/contrib/exec-smartctl

/usr/share/collectd/exec_hddpwrmode.sh

#!/bin/bash
HOSTNAME="${COLLECTD_HOSTNAME:-$(hostname -f)}"
INTERVAL="${COLLECTD_INTERVAL:-60}"

while sleep "$INTERVAL"
do
  for disk in sda sdb sdc sdd sde sdf
  do
    STATE=$(sudo smartctl -i -n standby /dev/$disk | grep -e "Device is in STANDBY mode" -e "Power mode is:    ACTIVE or IDLE" 2>/dev/null)
    if [ "$STATE" = "Device is in STANDBY mode, exit(2)" ]
      then
        # STANBY
        VALUE="0"
      else
        if [ "$STATE" = "Power mode is:    ACTIVE or IDLE" ]
          then
            # ACTIVE or IDLE
            VALUE="1"
          else
            # ERROR
            VALUE="U"
        fi
    fi
    echo "PUTVAL $HOSTNAME/disk-$disk/disk-state interval=$INTERVAL N:$VALUE" | tee -a /tmp/hddpwrstate.log
  done
done

The output in /tmp/hddpwrstate.log looks good. /tmp/hddpwrstate.log的输出看起来不错。

PUTVAL magneto/exec-smart/pwrstate_sdf interval=10.000 N:0
PUTVAL magneto/exec-smart/pwrstate_sda interval=10.000 N:0
PUTVAL magneto/exec-smart/pwrstate_sdb interval=10.000 N:1
PUTVAL magneto/exec-smart/pwrstate_sdc interval=10.000 N:0
PUTVAL magneto/exec-smart/pwrstate_sdd interval=10.000 N:0
PUTVAL magneto/exec-smart/pwrstate_sde interval=10.000 N:0
PUTVAL magneto/exec-smart/pwrstate_sdf interval=10.000 N:0

That means I have no permission issue (the script needs to be run by a user that can use sudo). 这意味着我没有权限问题(脚本必须由可以使用sudo的用户运行)。 And it is executed by collectd 并由收集执行

But I get no new rrd-files in /var/lib/collectd/rrd/<hostname> or new measurements in my InfluxDB. 但是我在/var/lib/collectd/rrd/<hostname>没有新的rrd文件,在我的InfluxDB中也没有新的度量。

> show measurements
name: measurements
name
----
cpu_value
df_value
disk_io_time
disk_read
disk_value
disk_weighted_io_time
disk_write
entropy_value
interface_rx
interface_tx
irq_value
load_longterm
load_midterm
load_shortterm
memory_value
processes_value
rrdcached_value
swap_value
uptime_value
users_value

I tried activating debug. 我尝试激活调试。

<Plugin syslog>
  LogLevel debug
</Plugin>

But it does not help. 但这无济于事。

Nov 19 12:13:12 magneto collectd[30028]: Exiting normally.
Nov 19 12:13:12 magneto collectd[30028]: collectd: Stopping 5 read threads.
Nov 19 12:13:12 magneto collectd[30028]: exec plugin: Sent SIGTERM to 30042
Nov 19 12:13:12 magneto collectd[30028]: collectd: Stopping 5 write threads.
Nov 19 12:13:13 magneto collectd[31752]: Stopping statistics collection and monitoring daemon: collectd.
Nov 19 12:13:13 magneto collectd[31767]: syslog: invalid loglevel [debug] defaulting to 'info'
Nov 19 12:13:13 magneto collectd[31768]: syslog: invalid loglevel [debug] defaulting to 'info'
Nov 19 12:13:13 magneto collectd[31769]: Initialization complete, entering read-loop.
Nov 19 12:13:13 magneto collectd[31764]: Starting statistics collection and monitoring daemon: collectd.

So, why does my script not work? 那么,为什么我的脚本不起作用? Is there something wrong? 有什么不对? Or how can I debug to find the issue myself? 或者我该如何调试自己找到问题?

My system: Debian 8 with OpenMediaVault. 我的系统:Debian 8和OpenMediaVault。

I got it working. 我知道了

Loks like my PUTVAL was not correct. 像我的PUTVAL这样的错误消息是不正确的。

This is working now: 现在正在工作:

echo "PUTVAL $HOSTNAME/exec-$disk/gauge-disk_state interval=$INTERVAL N:$VALUE" | tee -a /tmp/hddpwrstate.log

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM