Item: Zabbix
Rating: 10
Author: Verified User

Use Cases and Deployment Scope

Zabbix is used to monitor about 5000 network devices, 1000 servers and other services totaling 8000 hosts. Our IT department is deploying Zabbix to replace older monitoring services, to expand and modernize monitoring and alerting. Alerting is being automated and alerts are to be distributed to the responsible parties and even automate recovery where possible. Zabbix also stores a lot of useful "normal state" performance metrics for reference during problems.

Pros and Cons

Supports many different devices and server platforms
Customization of alert thresholds and notification actions
Automatic resolution of problems via remote commands

High level management always wants a "dashboard", while Zabbix can provide several different views, there seem to be third party products to provide alternatives. Maybe a best of breed dashboard could be added to the base product.
Zabbix performance largely depends on the performance of the underlying database, it takes planing and good infrastructure to support large environments.
It is possible to break things accidentally when making configuration changes, using Update when you meant to Clone, maybe a few strategic "are you sure you want to change this" prompts would help.

Return on Investment

A 5 year "Linux team only" implementation leveraged automatic recovery scripts to eliminate a huge number of off hour wake-up tickets. Many problems were detected and automatically remediated before the old monitoring system ever detected the problem.
The ability to monitor almost anything you can imagine and script.
Quick one-off items and triggers can be deployed in a crisis.
Leveraging Puppet and The Foreman for server build and Zabbix for monitoring made it possible for system administrators to go from supporting about 40 servers each to 120 or more.