Generating a Configuration File
A default Telegraf config file can be generated using the
telegraf -sample-config > telegraf.conf
To generate a file with specific inputs and outputs, you can use the
telegraf -sample-config -input-filter cpu:mem:net:swap -output-filter influxdb:kafka
You can see the latest config file with all available plugins here: telegraf.conf.
Environment variables can be used anywhere in the config file, simply prepend
For strings the variable must be within quotes (ie,
for numbers and booleans they should be plain (ie,
Global tags can be specified in the
[global_tags] section of the config file
in key=“value” format. All metrics being gathered on this host will be tagged
with the tags specified here.
Telegraf has a few options you can configure under the
agent section of the
- interval: Default data collection interval for all inputs
- round_interval: Rounds collection interval to ‘interval’ ie, if interval=“10s” then always collect on :00, :10, :20, etc.
- metric_batch_size: Telegraf will send metrics to output in batch of at most metric_batch_size metrics.
- metric_buffer_limit: Telegraf will cache metric_buffer_limit metrics for each output, and will flush this buffer on a successful write. This should be a multiple of metric_batch_size and could not be less than 2 times metric_batch_size.
- collection_jitter: Collection jitter is used to jitter the collection by a random amount. Each plugin will sleep for a random time within jitter before collecting. This can be used to avoid many plugins querying things like sysfs at the same time, which can have a measurable effect on the system.
- flush_interval: Default data flushing interval for all outputs. You should not set this below interval. Maximum flush_interval will be flush_interval + flush_jitter
- flush_jitter: Jitter the flush interval by a random amount. This is primarily to avoid large write spikes for users running a large number of telegraf instances. ie, a jitter of 5s and flush_interval 10s means flushes will happen every 10-15s.
- debug: Run telegraf in debug mode.
- quiet: Run telegraf in quiet mode.
- hostname: Override default hostname, if empty use os.Hostname().
Filters can be configured per input or output, see below for examples.
- namepass: An array of strings that is used to filter metrics generated by the current input. Each string in the array is tested as a glob match against measurement names and if it matches, the field is emitted.
- namedrop: The inverse of pass, if a measurement name matches, it is not emitted.
- fieldpass: An array of strings that is used to filter metrics generated by the current input. Each string in the array is tested as a glob match against field names and if it matches, the field is emitted. fieldpass is not available for outputs.
- fielddrop: The inverse of pass, if a field name matches, it is not emitted. fielddrop is not available for outputs.
- tagpass: tag names and arrays of strings that are used to filter measurements by the current input. Each string in the array is tested as a glob match against the tag name, and if it matches the measurement is emitted.
- tagdrop: The inverse of tagpass. If a tag matches, the measurement is not emitted. This is tested on measurements that have passed the tagpass test.
- tagexclude: tagexclude can be used to exclude a tag from measurement(s). As opposed to tagdrop, which will drop an entire measurement based on it’s tags, tagexclude simply strips the given tag keys from the measurement. This can be used on inputs & outputs, but it is recommended to be used on inputs, as it is more efficient to filter out tags at the ingestion point.
- taginclude: taginclude is the inverse of tagexclude. It will only include the tag keys in the final measurement.
Some configuration options are configurable per input:
- name_override: Override the base name of the measurement. (Default is the name of the input).
- name_prefix: Specifies a prefix to attach to the measurement name.
- name_suffix: Specifies a suffix to attach to the measurement name.
- tags: A map of tags to apply to a specific input’s measurements.
- interval: How often to gather this metric. Normal plugins use a single global interval, but if one particular input should be run less or more often, you can configure that here.
Input Configuration Examples
This is a full working config that will output CPU data to an InfluxDB instance
at 192.168.59.103:8086, tagging measurements with dc=“denver-1”. It will output
measurements at a 10s interval and will collect per-cpu data, dropping any
fields which begin with
[global_tags] dc = "denver-1" [agent] interval = "10s" # OUTPUTS [[outputs.influxdb]] url = "http://192.168.59.103:8086" # required. database = "telegraf" # required. precision = "s" # INPUTS [[inputs.cpu]] percpu = true totalcpu = false # filter all fields beginning with 'time_' fielddrop = ["time_*"]
Input Config: tagpass and tagdrop
[[inputs.cpu]] percpu = true totalcpu = false fielddrop = ["cpu_time"] # Don't collect CPU data for cpu6 & cpu7 [inputs.cpu.tagdrop] cpu = [ "cpu6", "cpu7" ] [[inputs.disk]] [inputs.disk.tagpass] # tagpass conditions are OR, not AND. # If the (filesystem is ext4 or xfs) OR (the path is /opt or /home) # then the metric passes fstype = [ "ext4", "xfs" ] # Globs can also be used on the tag values path = [ "/opt", "/home*" ]
Input Config: fieldpass and fielddrop
# Drop all metrics for guest & steal CPU usage [[inputs.cpu]] percpu = false totalcpu = true fielddrop = ["usage_guest", "usage_steal"] # Only store inode related metrics for disks [[inputs.disk]] fieldpass = ["inodes*"]
Input Config: namepass and namedrop
# Drop all metrics about containers for kubelet [[inputs.prometheus]] urls = ["http://kube-node-1:4194/metrics"] namedrop = ["container_*"] # Only store rest client related metrics for kubelet [[inputs.prometheus]] urls = ["http://kube-node-1:4194/metrics"] namepass = ["rest_client_*"]
Input Config: taginclude and tagexclude
# Only include the "cpu" tag in the measurements for the cpu plugin. [[inputs.cpu]] percpu = true totalcpu = true taginclude = ["cpu"] # Exclude the "fstype" tag from the measurements for the disk plugin. [[inputs.disk]] tagexclude = ["fstype"]
Input config: prefix, suffix, and override
This plugin will emit measurements with the name
[[inputs.cpu]] name_suffix = "_total" percpu = false totalcpu = true
This will emit measurements with the name
[[inputs.cpu]] name_override = "foobar" percpu = false totalcpu = true
Input config: tags
This plugin will emit measurements with two additional tags:
NOTE: Order matters, the
[inputs.cpu.tags] table must be at the end of the
[[inputs.cpu]] percpu = false totalcpu = true [inputs.cpu.tags] tag1 = "foo" tag2 = "bar"
Multiple inputs of the same type
Additional inputs (or outputs) of the same type can be specified,
just define more instances in the config file. It is highly recommended that
name_suffix config options
to avoid measurement collisions:
[[inputs.cpu]] percpu = false totalcpu = true [[inputs.cpu]] percpu = true totalcpu = false name_override = "percpu_usage" fielddrop = ["cpu_time*"]
Telegraf also supports specifying multiple output sinks to send data to,
configuring each output sink is different, but examples can be
found by running
[[outputs.influxdb]] urls = [ "http://localhost:8086" ] database = "telegraf" precision = "s" # Drop all measurements that start with "aerospike" namedrop = ["aerospike*"] [[outputs.influxdb]] urls = [ "http://localhost:8086" ] database = "telegraf-aerospike-data" precision = "s" # Only accept aerospike data: namepass = ["aerospike*"] [[outputs.influxdb]] urls = [ "http://localhost:8086" ] database = "telegraf-cpu0-data" precision = "s" # Only store measurements where the tag "cpu" matches the value "cpu0" [outputs.influxdb.tagpass] cpu = ["cpu0"]