The plugin-driven server agent for reporting metrics into InfluxDB...

Telegraf
1

21-07-21T23:39:04Z I! Starting Telegraf 1.19.1
21-07-21T23:39:04Z I! Loaded inputs: vsphere (3x)
21-07-21T23:39:04Z I! Loaded aggregators:
21-07-21T23:39:04Z I! Loaded processors:
21-07-21T23:39:04Z I! Loaded outputs: influxdb
21-07-21T23:39:04Z I! Tags enabled: host=Trunks-Dashboard
21-07-21T23:39:04Z I! [agent] Config: Interval:10s, Quiet:false, Hostname:“Trunks-Dashboard”, Flush Inte>
21-07-21T23:39:04Z I! [inputs.vsphere] Starting plugin
21-07-21T23:39:05Z I! [inputs.vsphere] Starting plugin
21-07-21T23:39:06Z I! [inputs.vsphere] Starting plugin
g The plugin-driven server agent for reporting metrics into InfluxDB…
21-07-22T00:06:19Z I! [inputs.vsphere] Stopping plugin
21-07-22T00:06:19Z I! [inputs.vsphere] Stopping plugin
21-07-22T00:06:19Z I! [inputs.vsphere] Stopping plugin
21-07-22T00:06:19Z I! [agent] Hang on, flushing any cached metrics before shutdown
21-07-22T00:06:19Z I! [agent] Stopping running outputs
f.service: Succeeded.
The plugin-driven server agent for reporting metrics into InfluxDB.
The plugin-driven server agent for reporting metrics into InfluxDB.
me=“2021-07-21T20:06:19-04:00” level=error msg="failed to create cache directory. /etc/telegraf/.cache/s>
me=“2021-07-21T20:06:19-04:00” level=error msg="failed to open. Ignored. open /etc/telegraf/.cache/snowf>
21-07-22T00:06:19Z I! Starting Telegraf 1.19.1
21-07-22T00:06:19Z I! Loaded inputs: vsphere (3x)
21-07-22T00:06:19Z I! Loaded aggregators:
21-07-22T00:06:19Z I! Loaded processors:
21-07-22T00:06:19Z I! Loaded outputs: influxdb
21-07-22T00:06:19Z I! Tags enabled: host=Trunks-Dashboard
21-07-22T00:06:19Z I! [agent] Config: Interval:10s, Quiet:false, Hostname:“Trunks-Dashboard”, Flush Inte>
21-07-22T00:06:19Z I! [inputs.vsphere] Starting plugin
21-07-22T00:06:20Z I! [inputs.vsphere] Starting plugin
21-07-22T00:06:20Z I! [inputs.vsphere] Starting plugin

I can run

/usr/bin/telegraf --config /etc/telegraf/telegraf.conf --config-directory /etc/telegraf/telegraf.d

Output:

2021-07-22T00:12:43Z I! Starting Telegraf 1.19.1
2021-07-22T00:12:43Z I! Loaded inputs: vsphere (3x)
2021-07-22T00:12:43Z I! Loaded aggregators:
2021-07-22T00:12:43Z I! Loaded processors:
2021-07-22T00:12:43Z I! Loaded outputs: influxdb
2021-07-22T00:12:43Z I! Tags enabled: host=Trunks-Dashboard
2021-07-22T00:12:43Z I! [agent] Config: Interval:10s, Quiet:false, Hostname:“Trunks-Dashboard”, Flush Interval:10s
2021-07-22T00:12:43Z I! [inputs.vsphere] Starting plugin
2021-07-22T00:12:44Z I! [inputs.vsphere] Starting plugin
2021-07-22T00:12:44Z I! [inputs.vsphere] Starting plugin
^C2021-07-22T00:16:26Z I! [inputs.vsphere] Stopping plugin
2021-07-22T00:16:26Z I! [inputs.vsphere] Stopping plugin
2021-07-22T00:16:26Z I! [inputs.vsphere] Stopping plugin
2021-07-22T00:16:26Z I! [agent] Hang on, flushing any cached metrics before shutdown
2021-07-22T00:16:26Z I! [agent] Stopping running outputs

I run:
/usr/bin/telegraf --config /etc/telegraf/telegraf.conf --config-directory /etc/telegraf/telegraf.d --test

output:
2021-07-22T00:16:40Z I! Starting Telegraf 1.19.1
2021-07-22T00:16:41Z I! [inputs.vsphere] Starting plugin
2021-07-22T00:16:41Z I! [inputs.vsphere] Starting plugin
2021-07-22T00:16:42Z I! [inputs.vsphere] Starting plugin

vsphere_datastore_disk,dcname=****** and so on

When I run it the second time with the test I get all of the output I would expect to go into influx from the ESXI host. I’m lost as to what the difference is. Any help would be appreciated and I’ll provide any info needed to help resolve this.

2

I didn’t get the question… what are you trying to do? what is your expectation?

3

if I run /usr/bin/telegraf --config /etc/telegraf/telegraf.conf --config-directory /etc/telegraf/telegraf.d --test

I get all the output i would expect from esxi. When I let telegraf run normally it closes and doesn’t give any output into influx.

4

The --test flag won’t write any data, it just executes inputs, processors, and aggregators, then sends all the data to the standard output (ie: console).

To keep fetching data you should register it as a service… or let it run in a console. If you want to do a one-shot gathering run you can use the flag --once.

5

@morgan9220
Do you have any outputs plugin configured in your telegraf.conf file?
You need at least one output plugin.
For debugging purposes you could also use the outputs.file plugin.
But you need at least one.