fix(telegraf) Document agent configuration

fixes #48

Author: Jonathan Chauncey

telegraf/README.md

@@ -11,6 +11,22 @@ You must do 2 things if you want to receive host level metrics from telegraf.
  * - Set the volume mounts for both `/sys` and `/proc`
  * - Set the environment variables `HOST_PROC` and `HOST_SYS` to the values of where `/sys` and `/proc` are mounted in the container. Example entries can be found in the [manifest](manifests/deis-monitor-telegraf-daemon.yaml).
 
+ ## Environment Variables
+ The configuration is driven via environment variables which are published to the `config.toml` file passed to telegraf when it starts. The following table gives the environment variable name and the default value if it is not set.
+
+ | Name | Default | Description |
+ |-----------|---------------|---------------|
+ | AGENT_INTERVAL | 10s | Default data collection interval for all inputs |
+ | AGENT_ROUND_INTERVAL | true | Rounds collection interval to 'interval' ie, if interval="10s" then always collect on :00, :10, :20, etc. |
+ | AGENT_BUFFER_LIMIT | 10000 | Telegraf will cache metric_buffer_limit metrics for each output, and will flush this buffer on a successful write. |
+ | AGENT_COLLECTION_JITTER | 0s | 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. |
+ | AGENT_FLUSH_INTERVAL | 10s | Default data flushing interval for all outputs. You should not set this below interval. Maximum flush_interval will be flush_interval + flush_jitter |
+ | AGENT_FLUSH_JITTER | 0s | 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. |
+ | AGENT_DEBUG | false | Run telegraf in debug mode. |
+ | AGENT_QUIET | false | Run telegraf in quiet mode. |
+ | AGENT_HOSTNAME | NodeName | Override default hostname |
+ 
+
 ## Development
 There is a make file provided with the project that can build the image, push it to a registry, and deploy it to a kubernetes cluster.
 

telegraf/rootfs/config.toml.tpl

@@ -3,96 +3,96 @@
 
 # Set Agent Configuration
 [agent]
-  interval = "{{ default "10s" .AGENT_INTERVAL}}"
-  round_interval = {{ default "true" .AGENT_ROUND_INTERVAL }}
+  interval = {{ default "10s" .AGENT_INTERVAL | quote }}
+  round_interval = {{ default true .AGENT_ROUND_INTERVAL }}
   metric_buffer_limit = {{ default "10000" .AGENT_BUFFER_LIMIT }}
-  collection_jitter = "{{ default "0s" .AGENT_COLLECTION_JITTER }}"
-  flush_interval = "{{ default "10s" .AGENT_FLUSH_INTERVAL }}"
-  flush_jitter = "{{ default "0s" .AGENT_FLUSH_JITTER }}"
-  debug = {{ default "false" .AGENT_DEBUG }}
-  quiet = {{ default "false" .AGENT_QUIET }}
-  {{ if .AGENT_HOSTNAME }}hostname = "{{ .AGENT_HOSTNAME }}"{{ end }}
+  collection_jitter = {{ default "0s" .AGENT_COLLECTION_JITTER | quote }}
+  flush_interval = {{ default "10s" .AGENT_FLUSH_INTERVAL | quote }}
+  flush_jitter = {{ default "0s" .AGENT_FLUSH_JITTER | quote }}
+  debug = {{ default false .AGENT_DEBUG }}
+  quiet = {{ default false .AGENT_QUIET }}
+  {{ if .AGENT_HOSTNAME }}hostname = {{ .AGENT_HOSTNAME | quote }} {{ end }}
 
 # Set output configuration
 {{ if .AMON_INSTANCE }}
 [[outputs.amon]]
-  server_key = "{{ .AMON_SERVER_KEY }}"
-  amon_instance = "{{ .AMON_INSTANCE }}"
-  timeout = "{{ default "5s" .AMON_TIMEOUT }}"
+  server_key = {{ .AMON_SERVER_KEY | quote }}
+  amon_instance = {{ .AMON_INSTANCE | quote }}
+  timeout = {{ default "5s" .AMON_TIMEOUT | quote }}
 {{ end }}
 
 {{ if .AMQP_URL }}
 [[outputs.amqp]]
-  url = "{{ .AMQP_URL }}"
-  exchange = "{{ default "telegraf" .AMQP_EXCHANGE }}"
-  routing_tag = "{{ default "host" .AMQP_ROUTING_TAG }}"
-  {{ if .AMQP_SSL_CA }} ssl_ca = "{{ .AMQP_SSL_CA }}" {{ end }}
-  {{ if .AMQP_SSL_CERT }} ssl_cert = "{{ .AMQP_SSL_CERT }}" {{ end }}
-  {{ if .AMQP_SSL_KEY }} ssl_key = "{{ .AMQP_SSL_KEY }}" {{ end }}
-  retention_policy = "{{ default "default" .AMQP_RETENTION_POLICY }}"
-  database = = "{{ default "telegraf" .AMQP_DATABASE }}"
-  precision = "{{ default "s" .AMQP_PRECISION }}"
+  url = {{ .AMQP_URL | quote }}
+  exchange = {{ default "telegraf" .AMQP_EXCHANGE | quote }}
+  routing_tag = {{ default "host" .AMQP_ROUTING_TAG | quote }}
+  {{ if .AMQP_SSL_CA }} ssl_ca = {{ .AMQP_SSL_CA | quote }} {{ end }}
+  {{ if .AMQP_SSL_CERT }} ssl_cert = {{ .AMQP_SSL_CERT | quote }} {{ end }}
+  {{ if .AMQP_SSL_KEY }} ssl_key = {{ .AMQP_SSL_KEY | quote }} {{ end }}
+  retention_policy = {{ default "default" .AMQP_RETENTION_POLICY | quote }}
+  database = = {{ default "telegraf" .AMQP_DATABASE | quote }}
+  precision = {{ default "s" .AMQP_PRECISION | quote }}
 {{ end }}
 
 {{ if .DATADOG_API_KEY }}
 [[outputs.datadog]]
-  apikey = "{{ .DATADOG_API_KEY }}"
+  apikey = {{ .DATADOG_API_KEY | quote }}
 {{ end }}
 
 {{ if .GRAPHITE_SERVERS }}
 [[outputs.graphite]]
   servers = [{{ .GRAPHITE_SERVERS }}]
-  prefix = "{{ default "\"\"" .GRAPHITE_PREFIX }}"
+  prefix = {{ default "" .GRAPHITE_PREFIX | quote }}
   timeout = {{ default 2 .GRAPHITE_TIMEOUT }}
 {{ end }}
 
 {{ if .INFLUXDB_URLS}}
 [[outputs.influxdb]]
-  urls = ["{{ .INFLUXDB_URLS}}"]
-  database = "{{default "telegraf" .INFLUXDB_DATABASE }}"
-  precision = "{{ default "s" .INFLUXDB_PRECISION }}"
-  timeout = "{{ default "5s" .INFLUXDB_TIMEOUT }}"
-  {{ if .INFLUXDB_USERNAME}} username = {{ .INFLUXDB_USERNAME }} {{ end }}
-  {{ if .INFLUXDB_PASSWORD}} password = {{ .INFLUXDB_PASSWORD }} {{ end }}
-  {{ if .INFLUXDB_USER_AGENT}} user_agent = {{ .INFLUXDB_USER_AGENT }} {{ end }}
-  {{ if .INFLUXDB_UDP_PAYLOAD}} udp_payload = {{ .INFLUXDB_UDP_PAYLOAD }} {{ end }}
+  urls = [{{ .INFLUXDB_URLS | quote }}]
+  database = {{default "telegraf" .INFLUXDB_DATABASE | quote }}
+  precision = {{ default "s" .INFLUXDB_PRECISION | quote }}
+  timeout = {{ default "5s" .INFLUXDB_TIMEOUT | quote }}
+  {{ if .INFLUXDB_USERNAME}} username = {{ .INFLUXDB_USERNAME | quote }} {{ end }}
+  {{ if .INFLUXDB_PASSWORD}} password = {{ .INFLUXDB_PASSWORD | quote }} {{ end }}
+  {{ if .INFLUXDB_USER_AGENT}} user_agent = {{ .INFLUXDB_USER_AGENT | quote }} {{ end }}
+  {{ if .INFLUXDB_UDP_PAYLOAD}} udp_payload = {{ .INFLUXDB_UDP_PAYLOAD | quote }} {{ end }}
 {{ end }}
 
 {{ if .KAFKA_BROKERS}}
 [[outputs.kafka]]
-  brokers = [{{ .KAFKA_BROKERS }}]
-  topic = "{{ default "telegraf" .KAFKA_TOPIC }}"
-  routing_tag = "{{ .KAFKA_ROUTING_TAG }}"
-  certificate = "{{ .KAFKA_CERTIFICATE }}"
-  key = "{{ .KAFKA_KEY }}"
-  ca = "{{ .KAFKA_CA }}"
-  verify_ssl = {{ .KAFKA_VERIFY_SSL }}
+  brokers = [{{ .KAFKA_BROKERS | quote }}]
+  topic = {{ default "telegraf" .KAFKA_TOPIC | quote }}
+  routing_tag = {{ .KAFKA_ROUTING_TAG | quote }}
+  {{ if .KAFKA_CERTIFICATE }} certificate = {{ .KAFKA_CERTIFICATE | quote }} {{ end }}
+  {{ if .KAFKA_KEY }} key = {{ .KAFKA_KEY | quote }} {{ end }}
+  {{ if .KAFKA_CA }} ca = {{ .KAFKA_CA | quote }} {{ end }}
+  {{ if .KAFKA_VERIFY_SSL }} verify_ssl = {{ .KAFKA_VERIFY_SSL }} {{ end }}
 {{ end }}
 
 {{ if .LIBRATO_API_TOKEN }}
 [[outputs.librato]]
-  api_user = "{{ .LIBRATO_API_USER }}"
-  api_token = "{{ .LIBRATO_API_TOKEN }}"
-  source_tag = "{{ .LIBRATO_SOURCE_TAG }}"
+  api_user = {{ .LIBRATO_API_USER | quote }}
+  api_token = {{ .LIBRATO_API_TOKEN | quote }}
+  source_tag = {{ .LIBRATO_SOURCE_TAG | quote }}
 {{ end }}
 
 {{ if .NSQ_SERVER }}
 [[outputs.nsq]]
-  server = "{{ .NSQ_SERVER }}"
-  topic = "{{ .NSQ_TOPIC }}"
+  server = {{ .NSQ_SERVER | quote }}
+  topic = {{ .NSQ_TOPIC | quote }}
 {{ end }}
 
 {{ if .OPEN_TSDB_HOST }}
-  prefix = "{{ .OPEN_TSDB_PREFIX }}"
-  host = "{{ .OPEN_TSDB_HOST }}"
+  prefix = {{ .OPEN_TSDB_PREFIX | quote }}
+  host = {{ .OPEN_TSDB_HOST | quote }}
   port = {{ .OPEN_TSDB_PORT }}
   debug = {{ .OPEN_TSDB_DEUBG }}
 {{ end }}
 
 {{ if .RIEMANN_URL }}
 [[outputs.riemann]]
-  url = "{{ .RIEMANN_URL }}"
-  transport = "{{ .RIEMANN_TRANSPORT }}"
+  url = {{ .RIEMANN_URL | quote }}
+  transport = {{ .RIEMANN_TRANSPORT | quote }}
 {{ end }}
 
 # Set Input Configuration
@@ -103,7 +103,6 @@
 [[inputs.cpu]]
   percpu = true
   totalcpu = true
-  drop = ["time_*"]
 
 {{ if .ENABLE_ZFS }}[[inputs.zfs]]{{ end }}
 
@@ -131,7 +130,7 @@
 
 {{ if .DOCKER_ENDPOINT }}
 [[inputs.docker]]
-  endpoint = "{{ .DOCKER_ENDPOINT }}"
+  endpoint = {{ .DOCKER_ENDPOINT | quote }}
   container_names = [{{ if .DOCKER_CONTAINER_NAMES }}{{ .DOCKER_CONTAINER_NAMES }}{{ end }}]
 {{ end }}
 
@@ -149,18 +148,18 @@
 
 {{ if .HTTP_JSON_SERVERS }}
 [[inputs.httpjson]]
-  name = "{{.HTTP_JSON_NAME}}"
-  servers = [{{.HTTP_JSON_SERVERS}}]
-  method = "{{.HTTP_JSON_METHOD}}"
-  tag_keys = [{{.HTTP_JSON_TAG_KEYS}}]
+  name = {{ .HTTP_JSON_NAME | quote }}
+  servers = [{{ .HTTP_JSON_SERVERS }}]
+  method = {{ .HTTP_JSON_METHOD | quote }}
+  tag_keys = [{{ .HTTP_JSON_TAG_KEYS }}]
   [json.parameters]
-    event_type = "{{.HTTP_JSON_EVENT_TYPE}}"
-    threshold = "{{.HTTP_JSON_THRESHOLD}}"
+    event_type = {{ .HTTP_JSON_EVENT_TYPE | quote }}
+    threshold = {{ .HTTP_JSON_THRESHOLD | quote }}
 {{end}}
 
 {{ if .INFLUXDB_URLS }}
 [[inputs.influxdb]]
-  urls = ["{{ .INFLUXDB_URLS }}/debug/vars"]
+  urls = [{{ .INFLUXDB_URLS }}/debug/vars"]
 {{ end }}
 
 {{ if .MEMCACHED_SERVERS }}
@@ -190,22 +189,22 @@
 
 {{ if .POSTGRESQL_ADDRESS }}
 [[inputs.postgresql]]
-  address = "{{ .POSTGRESQL_ADDRESS }}"
+  address = {{ .POSTGRESQL_ADDRESS | quote }}
   {{ if .POSTGRESQL_DATABASES }} databases = [{{ .POSTGRESQL_DATABASES }}]  {{ end }}
 {{ end }}
 
 {{ if .PROMETHEUS_URLS }}
 [[inputs.prometheus]]
   urls = [{{ .PROMETHEUS_URLS }}]
-  bearer_token = "{{ .PROMETHEUS_BEARER_TOKEN }}"
+  bearer_token = {{ .PROMETHEUS_BEARER_TOKEN | quote }}
 {{ end }}
 
 {{ if .RABBITMQ_URL }}
 [[inputs.rabbitmq]]
-  url = "{{ .RABBITMQ_URL }}"
-  {{ if .RABBITMQ_NAME}} name = "{{ .RABBITMQ_NAME }}" {{ end }}
-  {{ if .RABBITMQ_USERNAME }} username = "{{ .RABBITMQ_USERNAME }}" {{ end }}
-  {{ if .RABBITMQ_PASSWORD }} password = "{{ .RABBITMQ_PASSWORD }}" {{ end }}
+  url = {{ .RABBITMQ_URL | quote }}
+  {{ if .RABBITMQ_NAME}} name = {{ .RABBITMQ_NAME | quote }} {{ end }}
+  {{ if .RABBITMQ_USERNAME }} username = {{ .RABBITMQ_USERNAME | quote }} {{ end }}
+  {{ if .RABBITMQ_PASSWORD }} password = {{ .RABBITMQ_PASSWORD | quote }} {{ end }}
   {{ if .RABBITMQ_NODES }} nodes = [{{ .RABBITMQ_NODES}}] {{ end }}
 {{ end }}
 
@@ -229,22 +228,22 @@
 [[inputs.kafka_consumer]]
   topics = [{{ .KAFKA_CONSUMER_TOPICS }}]
   zookeeper_peers = [{{ .KAFKA_CONSUMER_ZOOKEEPER_PEERS }}]
-  consumer_group = "{{ .KAFKA_CONSUMER_GROUP }}"
+  consumer_group = {{ .KAFKA_CONSUMER_GROUP | quote }}
   point_buffer = {{ .KAFKA_CONSUMER_POINT_BUFFER }}
-  offset = "{{ .KAFKA_CONSUMER_OFFSET }}"
+  offset = {{ .KAFKA_CONSUMER_OFFSET | quote }}
 {{ end }}
 
 {{ if .STATSD_SERVICE_ADDRESS }}
 [[inputs.statsd]]
-  service_address = "{{ .STATSD_SERVICE_ADDRESS }}"
-  delete_gauges = {{ default "false" .STATSD_DELETE_GAUGES }}
-  delete_counters {{ default "false" .STATSD_DELETE_COUNTERS }}
-  delete_sets = {{ default "false" .STATSD_DELETE_SETS }}
-  delete_timings = {{ default "true" .STATSD_DELETE_TIMINGS}}
-  percentiles = [{{ default "90" .STATSD_PERCENTILES }}]
-  convert_names = {{ default "true" .STATSD_CONVERT_NAMES }}
+  service_address = {{ .STATSD_SERVICE_ADDRESS | quote }}
+  delete_gauges = {{ default false .STATSD_DELETE_GAUGES }}
+  delete_counters {{ default false .STATSD_DELETE_COUNTERS }}
+  delete_sets = {{ default false .STATSD_DELETE_SETS }}
+  delete_timings = {{ default true .STATSD_DELETE_TIMINGS}}
+  percentiles = [{{ default 90 .STATSD_PERCENTILES }}]
+  convert_names = {{ default true .STATSD_CONVERT_NAMES }}
   {{ if .STATSD_TEMPLATES}} templates = [{{ .STATSD_TEMPLATES }}] {{ end }}
-  allowed_pending_messages = {{ default "10000" .STATSD_ALLOWED_PENDING_MESSAGES }}
-  percentile_limit = {{ default "1000" .STATSD_PERCENTILE_LIMIT }}
-  udp_packet_size = {{ default "1500" .STATSD_UDP_PACKET_SIZE }}
+  allowed_pending_messages = {{ default 10000 .STATSD_ALLOWED_PENDING_MESSAGES }}
+  percentile_limit = {{ default 1000 .STATSD_PERCENTILE_LIMIT }}
+  udp_packet_size = {{ default 1500 .STATSD_UDP_PACKET_SIZE }}
 {{ end }}