v1.6.0 caps large integer values for SNMP metrics at 2147483647

[GainfulShrimp]

Bug report

After upgrading to v1.6.0, any values from SNMP input plugin larger than 2147483647 were reported as 2147483647 instead of their true value. I noticed this for network interface counter metrics.

Relevant telegraf.conf:

I haven't included whole config, but here's a snippet for getting 'if_counter' values from my SNMP-monitored hosts:

  [[inputs.snmp.table]]
    ## measurement name
    name = "if_counters"
    inherit_tags = [ "hostname" ]
    [[inputs.snmp.table.field]]
      name = "bytes_recv"
      oid = ".1.3.6.1.2.1.31.1.1.1.6"
    [[inputs.snmp.table.field]]
      name = "bytes_sent"
      oid = ".1.3.6.1.2.1.31.1.1.1.10"
    [[inputs.snmp.table.field]]
      name = "interface"
      oid = ".1.3.6.1.2.1.31.1.1.1.1"
      is_tag = true

System info:

Telegraf 1.6.0-1
System is a Raspberry Pi 3 (armhf) running latest Raspbian Stretch Lite

Steps to reproduce:

1. Setup Telegraf SNMP input plugin to monitor network interface 64-bit counters for busy interfaces.
2. Make sure some counter values are bigger than 2147483647 (e.g. using snmpget/snmpwalk to view the OID/MIB values you set up in step 1).
3. Add a file output for Telegraf so that you can easily see what's being sent to InfluxDB.

Expected behavior:

The true values from the interface counters should be reported, like this:

if_counters,interface=eth0,hostname=ubnt,agent_host=192.168.2.1,host=pi5 bytes_recv=624642439992i,bytes_sent=222558637140i 1524207960000000000
if_counters,interface=eth1,hostname=ubnt,agent_host=192.168.2.1,host=pi5 bytes_recv=225602092038i,bytes_sent=625001398888i 1524207960000000000

Actual behavior:

Values are all capped at 2147483647 before being sent:

if_counters,agent_host=192.168.2.1,host=pi5,hostname=ubnt,interface=eth0 bytes_recv=2147483647i,bytes_sent=2147483647i 1524207690000000000
if_counters,agent_host=192.168.2.1,host=pi5,hostname=ubnt,interface=eth1 bytes_recv=2147483647i,bytes_sent=2147483647i 1524207690000000000

Additional info:

This works fine in 1.5.3-1 (and many previous versions too).

Just to make sure it was 1.6.0-1 causing the issue, I downgraded Telegraf using the following commands (I kept the exact same config) and now the correct values are being sent again:

wget https://dl.influxdata.com/telegraf/releases/telegraf_1.5.3-1_armhf.deb
dpkg -i telegraf_1.5.3-1_armhf.deb

Use case: [Why is this important (helps with prioritizing requests)]

I'm not sure if this just affects SNMP or is a general issue with Telegraf and large (bigger than 32-bit) integers, but it feels quite important.