跳转至

NSQ

·

采集 NSQ 运行数据并以指标的方式上报到观测云。

配置

前置条件

推荐 NSQ 版本 >= 1.0.0，已测试的版本：

  • 1.2.1
  • 1.1.0
  • 0.3.8

采集器配置

进入 DataKit 安装目录下的 conf.d/nsq 目录，复制 nsq.conf.sample 并命名为 nsq.conf。示例如下：

[[inputs.nsq]]
  ## NSQ Lookupd HTTP API endpoint
  lookupd = "http://localhost:4161"

  ## NSQD HTTP API endpoint
  ## example:
  ##   ["http://localhost:4151"]
  nsqd = []

  ## time units are "ms", "s", "m", "h"
  interval = "10s"

  ## Set true to enable election
  election = true

  ## Optional TLS Config
  # tls_ca = "/etc/telegraf/ca.pem"
  # tls_cert = "/etc/telegraf/cert.pem"
  # tls_key = "/etc/telegraf/key.pem"
  ## Use TLS but skip chain & host verification
  # insecure_skip_verify = false

  [inputs.nsq.tags]
  # some_tag = "some_value"
  # more_tag = "some_other_value"

配置好后，重启 DataKit 即可。

NSQ 采集器提供两种配置方式，分别为 lookupd 和 nsqd
  • lookupd：配置 NSQ 集群的 lookupd 地址，采集器会自动发现 NSQ Server 并采集数据，扩展性更佳
  • nsqd：配置固定的 NSQ Daemon（nsqd）地址列表，采集器只会采集该列表的 NSQ Server 数据

以上两种配置方式是互斥的，lookupd 优先级更高，推荐使用 lookupd 配置方式。

目前可以通过 ConfigMap 方式注入采集器配置来开启采集器。

指标

以下所有数据采集，默认会追加名为 host 的全局 tag（tag 值为 DataKit 所在主机名），也可以在配置中通过 [inputs.nsq.tags] 指定其它标签：

 [inputs.nsq.tags]
  # some_tag = "some_value"
  # more_tag = "some_other_value"
  # ...

nsq_topics

Metrics of all topics in the NSQ cluster

  • 标签
Tag Description
channel Channel name
topic Topic name
  • 指标列表
Metric Description Type Unit
backend_depth Total number of unconsumed messages exceeding the max-queue-size. int count
deferred_count Number of messages that have been requeued and are not yet ready for re-sending. int count
depth Total number of unconsumed messages in the current channel. int count
in_flight_count Number of messages during the sending process or client processing that have not been sent FIN, REQ (requeued), or timed out. int count
message_count Total number of messages processed in the current channel. int count
requeue_count Number of messages that have timed out or have been sent REQ by the client. int count
timeout_count Number of messages that have timed out and are still unprocessed. int count

nsq_nodes

Metrics of all nodes in the NSQ cluster.

  • 标签
Tag Description
host Hostname
server_host Service address, that is host:ip.
  • 指标列表
Metric Description Type Unit
backend_depth Total number of unconsumed messages exceeding the max-queue-size. int count
depth Total number of unconsumed messages in the current node. int count
message_count Total number of messages processed by the current node. int count

文档评价

文档内容是否对您有帮助？ ×