Le Goût du Libre

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Monitoring Sensor Data in Zabbix ======

**Purpose:** Collect and monitor multiple values (e.g. ''Value1'', ''Value2'') from a custom script or sensor source.

**Pending review**

===== Integration Approaches =====

=== Option A: UserParameter via Zabbix Agent ===

**Use case:** Sensor is directly attached to the monitored host.

==== Setup ====

  * Create script at ''/usr/local/bin/fetch_values.sh'':
    <code>
    #!/bin/bash
    echo "$(get_value1)"   # Replace with actual readout logic
    echo "$(get_value2)"
    </code>

  * Make script executable:
    <code>
    chmod +x /usr/local/bin/fetch_values.sh
    </code>

  * Add UserParameters to agent config:
    Create or edit ''/etc/zabbix/zabbix_agentd.d/custom.conf''
    <code>
    UserParameter=value1,/usr/local/bin/fetch_values.sh | head -n1
    UserParameter=value2,/usr/local/bin/fetch_values.sh | tail -n1
    </code>

  * Restart agent:
    <code>
    sudo systemctl restart zabbix-agent
    </code>

  * In Zabbix frontend:
    Add items to your host:
      * Key: ''value1'', Type: Zabbix agent
      * Key: ''value2'', Type: Zabbix agent

**Pros:** Simple, native agent integration  
**Cons:** Requires sensor + agent on same host

---

=== Option B: zabbix_sender (Push-Based Monitoring) ===

**Use case:** Sensor runs externally or independently from agent.

==== Setup ====

  * Create push script:
    <code>
    echo "<ZABBIX_HOSTNAME> value1 $(get_value1)" > /tmp/zbxdata.txt
    echo "<ZABBIX_HOSTNAME> value2 $(get_value2)" >> /tmp/zbxdata.txt
    </code>

  * Define trapper items in Zabbix frontend:
    * Type: ''Zabbix trapper''
    * Keys: ''value1'', ''value2''
    * Host: ''<ZABBIX_HOSTNAME>''

  * Send to server:
    <code>
    zabbix_sender -z <ZABBIX_SERVER_IP> -i /tmp/zbxdata.txt
    </code>

**Pros:** Flexible, works from remote systems  
**Cons:** Requires trapper item configuration

---

===== Best Practices =====

=== Reliability ===
  * Add retries and error checks to scripts
  * Log sensor readouts and sender results
  * Validate data before transmission

=== Automation ===
  * Schedule sensor polling using ''cron'' or ''systemd timers''

=== Alerting ===
  * Use Zabbix triggers for:
    * High temperature / abnormal values
    * Missing updates (e.g. last value > 10 min old)

=== Scalability ===
  * Template monitoring logic for similar hosts
  * Use descriptive keys (e.g. ''sensor.zone1.value1'')