You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
yzl 93958d0fb0
zabbix6.0
1 year ago
..
README.md zabbix6.0 1 year ago
template_app_apache_agent.yaml zabbix6.0 1 year ago

README.md

Apache by Zabbix agent

Overview

This template is designed for the effortless deployment of Apache monitoring by Zabbix via Zabbix agent and doesn't require any external scripts. The template Apache by Zabbix agent - collects metrics by polling mod_status locally with Zabbix agent:

127.0.0.1
ServerVersion: Apache/2.4.41 (Unix)
ServerMPM: event
Server Built: Aug 14 2019 00:35:10
CurrentTime: Friday, 16-Aug-2019 12:38:40 UTC
RestartTime: Wednesday, 14-Aug-2019 07:58:26 UTC
ParentServerConfigGeneration: 1
ParentServerMPMGeneration: 0
ServerUptimeSeconds: 189613
ServerUptime: 2 days 4 hours 40 minutes 13 seconds
Load1: 4.60
Load5: 1.20
Load15: 0.47
Total Accesses: 27860
Total kBytes: 33011
Total Duration: 54118
CPUUser: 18.02
CPUSystem: 31.76
CPUChildrenUser: 0
CPUChildrenSystem: 0
CPULoad: .0262535
Uptime: 189613
ReqPerSec: .146931
BytesPerSec: 178.275
BytesPerReq: 1213.33
DurationPerReq: 1.9425
BusyWorkers: 7
IdleWorkers: 93
Processes: 4
Stopping: 0
BusyWorkers: 7
IdleWorkers: 93
ConnsTotal: 13
ConnsAsyncWriting: 0
ConnsAsyncKeepAlive: 5
ConnsAsyncClosing: 0
Scoreboard: ...

It also uses Zabbix agent to collect Apache Linux process statistics such as CPU usage, memory usage, and whether the process is running or not.

Requirements

Zabbix version: 7.0 and higher.

Tested versions

This template has been tested on:

  • Apache 2.4.41

Configuration

Zabbix should be configured according to the instructions in the Templates out of the box section.

Setup

See the setup instructions for mod_status.

Check the availability of the module with this command line: httpd -M 2>/dev/null | grep status_module

This is an example configuration of the Apache web server:

<Location "/server-status">
  SetHandler server-status
  Require host example.com
</Location>

If you use another path, then do not forget to change the {$APACHE.STATUS.PATH} macro. Install and setup Zabbix agent.

Macros used

Name Description Default
{$APACHE.STATUS.HOST}

The hostname or IP address of the Apache status page.

127.0.0.1
{$APACHE.STATUS.PORT}

The port of the Apache status page.

80
{$APACHE.STATUS.PATH}

The URL path.

server-status?auto
{$APACHE.STATUS.SCHEME}

The request scheme, which may be either HTTP or HTTPS.

http
{$APACHE.RESPONSE_TIME.MAX.WARN}

The maximum Apache response time expressed in seconds for a trigger expression.

10
{$APACHE.PROCESS_NAME}

The process name filter for the Apache process discovery.

(httpd|apache2)
{$APACHE.PROCESS.NAME.PARAMETER}

The process name of the Apache web server used in the item key proc.get. It could be specified if the correct process name is known.

Items

Name Description Type Key and additional info
Apache: Get status

Getting data from a machine-readable version of the Apache status page.

For more information see Apache Module mod_status.

Zabbix agent web.page.get["{$APACHE.STATUS.SCHEME}://{$APACHE.STATUS.HOST}:{$APACHE.STATUS.PORT}/{$APACHE.STATUS.PATH}"]

Preprocessing

  • JavaScript: The text is too long. Please see the template.

Apache: Service ping Zabbix agent net.tcp.service[http,"{$APACHE.STATUS.HOST}","{$APACHE.STATUS.PORT}"]

Preprocessing

  • Discard unchanged with heartbeat: 10m

Apache: Service response time Zabbix agent net.tcp.service.perf[http,"{$APACHE.STATUS.HOST}","{$APACHE.STATUS.PORT}"]
Apache: Total bytes

The total bytes served.

Dependent item apache.bytes

Preprocessing

  • JSON Path: $["Total kBytes"]

  • Custom multiplier: 1024

Apache: Bytes per second

It is calculated as a rate of change for total bytes statistics.

BytesPerSec is not used, as it counts the average since the last Apache server start.

Dependent item apache.bytes.rate

Preprocessing

  • JSON Path: $["Total kBytes"]

  • Custom multiplier: 1024

  • Change per second
Apache: Requests per second

It is calculated as a rate of change for the "Total requests" statistics.

ReqPerSec is not used, as it counts the average since the last Apache server start.

Dependent item apache.requests.rate

Preprocessing

  • JSON Path: $["Total Accesses"]

  • Change per second
Apache: Total requests

The total number of the Apache server accesses.

Dependent item apache.requests

Preprocessing

  • JSON Path: $["Total Accesses"]

Apache: Uptime

The service uptime expressed in seconds.

Dependent item apache.uptime

Preprocessing

  • JSON Path: $.ServerUptimeSeconds

Apache: Version

The Apache service version.

Dependent item apache.version

Preprocessing

  • JSON Path: $.ServerVersion

  • Discard unchanged with heartbeat: 1d

Apache: Total workers busy

The total number of busy worker threads/processes.

Dependent item apache.workers_total.busy

Preprocessing

  • JSON Path: $.BusyWorkers

Apache: Total workers idle

The total number of idle worker threads/processes.

Dependent item apache.workers_total.idle

Preprocessing

  • JSON Path: $.IdleWorkers

Apache: Workers closing connection

The number of workers in closing state.

Dependent item apache.workers.closing

Preprocessing

  • JSON Path: $.Workers.closing

Apache: Workers DNS lookup

The number of workers in dnslookup state.

Dependent item apache.workers.dnslookup

Preprocessing

  • JSON Path: $.Workers.dnslookup

Apache: Workers finishing

The number of workers in finishing state.

Dependent item apache.workers.finishing

Preprocessing

  • JSON Path: $.Workers.finishing

Apache: Workers idle cleanup

The number of workers in cleanup state.

Dependent item apache.workers.cleanup

Preprocessing

  • JSON Path: $.Workers.cleanup

Apache: Workers keepalive (read)

The number of workers in keepalive state.

Dependent item apache.workers.keepalive

Preprocessing

  • JSON Path: $.Workers.keepalive

Apache: Workers logging

The number of workers in logging state.

Dependent item apache.workers.logging

Preprocessing

  • JSON Path: $.Workers.logging

Apache: Workers reading request

The number of workers in reading state.

Dependent item apache.workers.reading

Preprocessing

  • JSON Path: $.Workers.reading

Apache: Workers sending reply

The number of workers in sending state.

Dependent item apache.workers.sending

Preprocessing

  • JSON Path: $.Workers.sending

Apache: Workers slot with no current process

The number of slots with no current process.

Dependent item apache.workers.slot

Preprocessing

  • JSON Path: $.Workers.slot

Apache: Workers starting up

The number of workers in starting state.

Dependent item apache.workers.starting

Preprocessing

  • JSON Path: $.Workers.starting

Apache: Workers waiting for connection

The number of workers in waiting state.

Dependent item apache.workers.waiting

Preprocessing

  • JSON Path: $.Workers.waiting

Apache: Get processes summary

The aggregated data of summary metrics for all processes.

Zabbix agent proc.get[{$APACHE.PROCESS.NAME.PARAMETER},,,summary]

Triggers

Name Description Expression Severity Dependencies and additional info
Apache: Host has been restarted

Uptime is less than 10 minutes.

last(/Apache by Zabbix agent/apache.uptime)<10m Info Manual close: Yes
Apache: Version has changed

Apache version has changed. Acknowledge to close the problem manually.

last(/Apache by Zabbix agent/apache.version,#1)<>last(/Apache by Zabbix agent/apache.version,#2) and length(last(/Apache by Zabbix agent/apache.version))>0 Info Manual close: Yes

LLD rule Event MPM discovery

Name Description Type Key and additional info
Event MPM discovery

The discovery of additional metrics if the event Multi-Processing Module (MPM) is used.

For more details see Apache MPM event.

Dependent item apache.mpm.event.discovery

Preprocessing

  • JavaScript: The text is too long. Please see the template.

  • Discard unchanged with heartbeat: 3h

Item prototypes for Event MPM discovery

Name Description Type Key and additional info
Apache: Connections async closing

The number of asynchronous connections in closing state (applicable only to the event MPM).

Dependent item apache.connections[async_closing{#SINGLETON}]

Preprocessing

  • JSON Path: $.ConnsAsyncClosing

Apache: Connections async keepalive

The number of asynchronous connections in keepalive state (applicable only to the event MPM).

Dependent item apache.connections[async_keep_alive{#SINGLETON}]

Preprocessing

  • JSON Path: $.ConnsAsyncKeepAlive

Apache: Connections async writing

The number of asynchronous connections in writing state (applicable only to the event MPM).

Dependent item apache.connections[async_writing{#SINGLETON}]

Preprocessing

  • JSON Path: $.ConnsAsyncWriting

Apache: Connections total

The number of total connections.

Dependent item apache.connections[total{#SINGLETON}]

Preprocessing

  • JSON Path: $.ConnsTotal

Apache: Bytes per request

The average number of client requests per second.

Dependent item apache.bytes[per_request{#SINGLETON}]

Preprocessing

  • JSON Path: $.BytesPerReq

Apache: Number of async processes

The number of asynchronous processes.

Dependent item apache.process[num{#SINGLETON}]

Preprocessing

  • JSON Path: $.Processes

LLD rule Apache process discovery

Name Description Type Key and additional info
Apache process discovery

The discovery of the Apache process summary.

Dependent item apache.proc.discovery

Item prototypes for Apache process discovery

Name Description Type Key and additional info
Apache: CPU utilization

The percentage of the CPU utilization by a process {#APACHE.NAME}.

Zabbix agent proc.cpu.util[{#APACHE.NAME}]
Apache: Get process data

The summary metrics aggregated by a process {#APACHE.NAME}.

Dependent item apache.proc.get[{#APACHE.NAME}]

Preprocessing

  • JSON Path: $.[?(@["name"]=="{#APACHE.NAME}")].first()

    Custom on fail: Set value to: Failed to retrieve process {#APACHE.NAME} data

Apache: Memory usage (rss)

The summary of resident set size memory used by a process {#APACHE.NAME} expressed in bytes.

Dependent item apache.proc.rss[{#APACHE.NAME}]

Preprocessing

  • JSON Path: $.rss

    Custom on fail: Discard value

Apache: Memory usage (vsize)

The summary of virtual memory used by a process {#APACHE.NAME} expressed in bytes.

Dependent item apache.proc.vmem[{#APACHE.NAME}]

Preprocessing

  • JSON Path: $.vsize

    Custom on fail: Discard value

Apache: Memory usage, %

The percentage of real memory used by a process {#APACHE.NAME}.

Dependent item apache.proc.pmem[{#APACHE.NAME}]

Preprocessing

  • JSON Path: $.pmem

    Custom on fail: Discard value

Apache: Number of running processes

The number of running processes {#APACHE.NAME}.

Dependent item apache.proc.num[{#APACHE.NAME}]

Preprocessing

  • JSON Path: $.processes

    Custom on fail: Set value to: 0

  • Discard unchanged with heartbeat: 1h

Trigger prototypes for Apache process discovery

Name Description Expression Severity Dependencies and additional info
Apache: Process is not running last(/Apache by Zabbix agent/apache.proc.num[{#APACHE.NAME}])=0 High
Apache: Service is down last(/Apache by Zabbix agent/net.tcp.service[http,"{$APACHE.STATUS.HOST}","{$APACHE.STATUS.PORT}"])=0 and last(/Apache by Zabbix agent/apache.proc.num[{#APACHE.NAME}])>0 Average Manual close: Yes
Apache: Failed to fetch status page

Zabbix has not received any data for items for the last 30 minutes.

nodata(/Apache by Zabbix agent/web.page.get["{$APACHE.STATUS.SCHEME}://{$APACHE.STATUS.HOST}:{$APACHE.STATUS.PORT}/{$APACHE.STATUS.PATH}"],30m)=1 and last(/Apache by Zabbix agent/apache.proc.num[{#APACHE.NAME}])>0 Warning Manual close: Yes
Depends on:
  • Apache: Service is down
Apache: Service response time is too high min(/Apache by Zabbix agent/net.tcp.service.perf[http,"{$APACHE.STATUS.HOST}","{$APACHE.STATUS.PORT}"],5m)>{$APACHE.RESPONSE_TIME.MAX.WARN} and last(/Apache by Zabbix agent/apache.proc.num[{#APACHE.NAME}])>0 Warning Manual close: Yes
Depends on:
  • Apache: Service is down

Feedback

Please report any issues with the template at https://support.zabbix.com

You can also provide feedback, discuss the template, or ask for help at ZABBIX forums