phpfpm

README

Telegraf plugin: phpfpm

Get phpfpm stat using either HTTP status page or fpm socket.

Measurements

Meta:

• tags: pool=poolname

Measurement names:

• phpfpm

Measurement field:

• accepted_conn
• listen_queue
• max_listen_queue
• listen_queue_len
• idle_processes
• active_processes
• total_processes
• max_active_processes
• max_children_reached
• slow_requests

Example output

Using this configuration:

[phpfpm]
  # An array of address to gather stats about. Specify an ip on hostname
  # with optional port and path. ie localhost, 10.10.3.33/server-status, etc.
  #
  # We can configure in three modes:
  #   - unixsocket: the string is the path to fpm socket like
  #      /var/run/php5-fpm.sock
  #   - http: the URL has to start with http:// or https://
  #   - fcgi: the URL has to start with fcgi:// or cgi://, and socket port must present
  #
  # If no servers are specified, then default to 127.0.0.1/server-status
  urls = ["http://localhost/status", "10.0.0.12:/var/run/php5-fpm-www2.sock", "fcgi://10.0.0.12:9000/status"]

When run with:

./telegraf -config telegraf.conf -input-filter phpfpm -test

It produces:

* Plugin: phpfpm, Collection 1
> phpfpm,pool=www accepted_conn=13i,active_processes=2i,idle_processes=1i,listen_queue=0i,listen_queue_len=0i,max_active_processes=2i,max_children_reached=0i,max_listen_queue=0i,slow_requests=0i,total_processes=3i 1453011293083331187
> phpfpm,pool=www2 accepted_conn=12i,active_processes=1i,idle_processes=2i,listen_queue=0i,listen_queue_len=0i,max_active_processes=2i,max_children_reached=0i,max_listen_queue=0i,slow_requests=0i,total_processes=3i 1453011293083691422
> phpfpm,pool=www3 accepted_conn=11i,active_processes=1i,idle_processes=2i,listen_queue=0i,listen_queue_len=0i,max_active_processes=2i,max_children_reached=0i,max_listen_queue=0i,slow_requests=0i,total_processes=3i 1453011293083691658

Note

When using unixsocket, you have to ensure that telegraf runs on same host, and socket path is accessible to telegraf user.

Documentation

Overview

Package fcgi implements the FastCGI protocol. Currently only the responder role is supported. The protocol is defined at http://www.fastcgi.com/drupal/node/6?q=node/22

Index

• Constants
• Variables
• func Serve(l net.Listener, handler http.Handler) error

Constants

const (
	PF_POOL                 = "pool"
	PF_PROCESS_MANAGER      = "process manager"
	PF_ACCEPTED_CONN        = "accepted conn"
	PF_LISTEN_QUEUE         = "listen queue"
	PF_MAX_LISTEN_QUEUE     = "max listen queue"
	PF_LISTEN_QUEUE_LEN     = "listen queue len"
	PF_IDLE_PROCESSES       = "idle processes"
	PF_ACTIVE_PROCESSES     = "active processes"
	PF_TOTAL_PROCESSES      = "total processes"
	PF_MAX_ACTIVE_PROCESSES = "max active processes"
	PF_MAX_CHILDREN_REACHED = "max children reached"
	PF_SLOW_REQUESTS        = "slow requests"
)

Variables

var ErrConnClosed = errors.New("fcgi: connection to web server closed")

ErrConnClosed is returned by Read when a handler attempts to read the body of a request after the connection to the web server has been closed.

var ErrRequestAborted = errors.New("fcgi: request aborted by web server")

ErrRequestAborted is returned by Read when a handler attempts to read the body of a request that has been aborted by the web server.

Functions

func Serve

func Serve(l net.Listener, handler http.Handler) error

Serve accepts incoming FastCGI connections on the listener l, creating a new goroutine for each. The goroutine reads requests and then calls handler to reply to them. If l is nil, Serve accepts connections from os.Stdin. If handler is nil, http.DefaultServeMux is used.