collectd-apache.md

collectd/apache

Monitor Type: collectd/apache (Source)

Accepts Endpoints: Yes

Multiple Instances Allowed: Yes

Overview

Monitors Apache webservice instances using the information provided by mod_status.

Worker states

Apache worker threads can be in one of the following states:

State	Remark
Open	Open (unused) slot - no process
Waiting	Idle and waiting for request
Sending	Serving response
KeepAlive	Kept alive for possible next request
Idle_cleanup	Idle and marked for cleanup
Closing	Closing connection
Logging	Writing to log file
Reading	Reading request
Finishing	Finishing as part of graceful shutdown
Starting	Starting up to serve

Apache Setup

To configure the Apache webserver itself to expose status metrics:

1. Enable the mod_status module in your Apache server.

2. Add the following configuration to your Apache server:

    ExtendedStatus on
    <Location /mod_status>
    SetHandler server-status
    </Location>
   

3. Restart Apache.

Note: Make sure that the URL you provide for your mod_status module ends in ?auto. This returns the status page as text/plain, which this plugin requires.

Config Examples

monitors:
 - type: collectd/apache
   host: localhost
   port: 80

If mod_status is exposed on an endpoint other than /mod_status, you can use the url config option to specify the path:

monitors:
 - type: collectd/apache
   host: localhost
   port: 80
   url: "http://{{.Host}}:{{.Port}}/server-status?auto"

For a full list of options, see Configuration.

Configuration

To activate this monitor in the Smart Agent, add the following to your agent config:

monitors:  # All monitor config goes under this key
 - type: collectd/apache
   ...  # Additional config

For a list of monitor options that are common to all monitors, see Common Configuration.

Config option	Required	Type	Description
host	yes	string	The hostname of the Apache server
port	yes	integer	The port number of the Apache server
name	no	string	This will be sent as the plugin_instance dimension and can be any name you like.
url	no	string	The URL, either a final URL or a Go template that will be populated with the host and port values. (default: http://{{.Host}}:{{.Port}}/mod_status?auto)
username	no	string
password	no	string

Metrics

These are the metrics available for this monitor. Metrics that are categorized as container/host (default) are in bold and italics in the list below.

• apache_bytes (cumulative)
  Amount of data served by Apache, in bytes.

• apache_connections (gauge)
  The number of connections that are being served by Apache. This is also equal to the number of busy worker threads, where 'busy' means any worker thread which has been started successfully and is not slated for idle cleanup.

• apache_idle_workers (gauge)
  The number of Apache workers that are idling. If this number is consistently low, then your server may be too busy and you may have to increase the number of threads. If it is consistently high, then the system may be under-utilized.

• apache_requests (cumulative)
  The number of requests that have been served by Apache. This metric is useful to know total requests and the rate at which Apache is able to serve them.

• apache_scoreboard.closing (gauge)
  This metric shows how many worker threads are in the process of closing TCP connections after serving a response. If this number is consistently high, then there might be a network issue or errant client preventing TCP tear-down.

• apache_scoreboard.dnslookup (gauge)
  This metric counts the number of worker threads that are performing a DNS lookup. If this number is too high, check if there is a DNS resolution problem at your server. This can affect Apache server performance.

• apache_scoreboard.finishing (gauge)
  The number of worker threads that are finishing as part of graceful server shutdown.

• apache_scoreboard.idle_cleanup (gauge)
  The number of worker threads that are idle and ready for clean-up.

• apache_scoreboard.keepalive (gauge)
  The number of worker threads that are maintaining keep-alive connections: keeping the connection "alive" after serving a response, in the expectation that another HTTP request will come on the same connection. At the end of the keep-alive interval, the connection is closed.

• apache_scoreboard.logging (gauge)
  This metric shows how many worker threads are busy writing to the log file. If this number is consistently high, your logging level may be too high or one or more modules may be too verbose.

• apache_scoreboard.open (gauge)
  This metric shows how many worker slots are open. The slots do not have a worker thread yet, but they can be spun up based on incoming requests.

• apache_scoreboard.reading (gauge)
  This metric shows how many workers are in the process of receiving requests (headers or body). If this number is consistently high, clients may be sending large headers or uploading large files.

• apache_scoreboard.sending (gauge)
  This metric shows how many workers are sending responses. It is normal for this to be a large number when measuring sites that serve large downloads.

• apache_scoreboard.starting (gauge)
  This metric shows how many workers are being started up. If this number is consistently high, then the system may be overloaded.

• apache_scoreboard.waiting (gauge)
  This metric shows how many worker threads are ready and waiting for requests to come in.

Non-default metrics (version 4.7.0+)

To emit metrics that are not default, you can add those metrics in the generic monitor-level extraMetrics config option. Metrics that are derived from specific configuration options that do not appear in the above list of metrics do not need to be added to extraMetrics.

To see a list of metrics that will be emitted you can run agent-status monitors after configuring this monitor in a running agent instance.

Dimensions

The following dimensions may occur on metrics emitted by this monitor. Some dimensions may be specific to certain metrics.

Name	Description
plugin_instance	Set to whatever you set in the name config option.