ping plugin is broken after introducing CombinedOutputTimeout for exec command runners

[alekiv]

Bug report

Looks like after implementing CombinedOutputTimeout for exec command runners ping command got killed before it timeouts internally.

System info:

telegraf v0.13.1
OS: Debian Jessie
192.168.101.6 destination is not accessible for pings

Steps to reproduce:

Configure ping plugin as follows:

[[inputs.ping]]

NOTE: this plugin forks the ping command. You may need to set capabilities

via setcap cap_net_raw+p /bin/ping

urls to ping

urls = [ "192.168.101.6" ]# required

number of pings to send (ping -c )

count = 5 # required

interval, in s, at which to ping. 0 == default (ping -i <PING_INTERVAL>)

ping_interval = 0.0

ping timeout, in s. 0 == no timeout (ping -t )

timeout = 5.0

interface to send ping from (ping -I )

#interface = ""

Expected behavior:

ping plugin detects that 192.168.101.6 is not accessible and sends data to influxdb

Actual behavior:

telegraf kills plugin due to internal timeout, which probably equals with timout configured under plugin settings and doesn't take into account number of pings (eg. count = 5)
telegraf log:
Fatal error processing ping output
, Command timed out.

After ping count change from 5 to 1, telegraf doesn't kill ping exec by timeout.

[sparrc]

yep, see pr #1305