Page 1 of 1
Monitor aws disk I/O
Posted: Mon Dec 15, 2014 3:51 am
by gopukrishnantec
I have AWS SSD drives with provisioned I/O of 1050 IOPS. How can I monitor the same with nagios when it reaches 90% of its I/O capacity ?
You guys are always helpful. Thanks in advance.
Re: Monitor aws disk I/O
Posted: Mon Dec 15, 2014 1:14 pm
by abrist
I would presume it is virtualized on top of the disks. You could try to check for io wait . . . .
Re: Monitor aws disk I/O
Posted: Tue Dec 16, 2014 8:32 am
by gopukrishnantec
I have found a plugin for the same but it is asking the below values which I doubt for mine (1050 IOPS):
IO Check Mode
Total Transfers/sec :
Read IO/Sec :
Write IO/Sec :
Bytes Read/Sec :
Bytes Written/Sec :
Queue Mode
Average size of requests:
Queue length of requests:
Wait Time Mode
Avg I/O Wait Time (ms):
Avg Read Wait Time (ms):
Avg Write Wait Time (ms):
Avg Service Wait Time (ms):
Avg CPU Utilization:
Please provide me the average values or the links if any other plugins are available.
Thank you all
Re: Monitor aws disk I/O
Posted: Tue Dec 16, 2014 8:54 am
by eloyd
I would write a module that checked iostat as follows:
Code: Select all
Note: Usage would be check_iostat -d DISK -w warn -c crit
#!/bin/sh
device=""
warn=""
crit=""
while [ -n "$1" ]; do
case "$1" in
-d) device="$2"; shift 2;;
-w) warn="$2"; shift 2;;
-c) crit="$2"; shift 2;;
*) shift 1;;
esac
done
tps=`iostat -d $device | tail -2 | head -1 | awk '{print $2}'`
if [ "$tps" -ge "$crit" ]; then
echo "CRITICAL - $device has $tps transactions per second, critical value is $crit"
exit 2
elif [ "$tps" -ge "$warn" ]; then
echo "WARNING - $device has $tps transactions per second, warning value is $warn"
exit 1
else
echo "OK - $device has $tps transactions per second"
exit 0
fi
Example would be:
Code: Select all
check_iostat -d /dev/sda -w 500 -c 1000
Re: Monitor aws disk I/O
Posted: Tue Dec 16, 2014 5:58 pm
by abrist
Thanks for the input Eric.