Zabbix - A must have tool in your infrastructure
March 28, 2017

Zabbix - A must have tool in your infrastructure

Deepshikha Gandhi | TrustRadius Reviewer
Score 7 out of 10
Vetted Review
Verified User

Overall Satisfaction with Zabbix

Our operations team uses Zabbix to monitor our company's physical and virtual infrastructure. Zabbix monitors our servers, network gear, CPU performance data and application uptime. We have integrated Zabbix alerts with PagerDuty to get pinged when an issue arises.
  • Monitoring performance indicators like CPU, memory, network, disk space and processes can be done easily with Zabbix agent, which is available for Linux, UNIX and Windows platforms.
  • Zabbix can gather stats like disk failures, temperature and voltage from hardware through IPMI, thus ensuring uptime and reduces risk.
  • Zabbix lets you integrate it with your custom checks. You can write your own check scripts in Bash, Python or Perl and integrate it with Zabbix.
  • Zabbix has a steep learning curve and doesn't have a very intuitive and user-friendly interface.
  • Zabbix is resource hungry. It uses a DB to store all the stats and configuration and this can grow exponentially depending on the number of hosts you are monitoring.
  • Zabbix doesn't have a very thorough documentation, so you have to search for issues and ask the Zabbix community at times.
  • By integrating Zabbix with Rundeck, we have been able to auto-resolve issues in our infrastructure which means lesser on-call alerts.
  • Zabbix and PagerDuty let us response to business critical issues instantly.
  • Zabbix dashboards and screens let us get an insight into the status of our infrastructure visually.
Nagios has some advantages over Zabbix like "flapping" detection and multiple alert levels - Error, Warning and OK.
However, the disadvantages of Nagios like needing an addon (NRPE) to monitor remote system internals (open files, running processes, memory, etc), no charting of monitored values (different systems like "Cacti" or "Nagiosgraph" can be bolted on) and the web interface being read only, we decided to go for Zabbix because of the extra features it provides when stacked up against Nagios.
Zabbix is great for monitoring protocols like HTTP, FTP, SSH, SMTP, SNMP, etc. It also has great features like visual analysis, customizable dashboards and system "templates". Features like auto-discovery and auto-registration let us manage infrastructure on the fly. The auto-registration function is very handy for automatic monitoring of a new AWS host. When a new cloud host is spun up, Zabbix will automatically start collecting performance and availability data of that node. On the other hand, Zabbix tuning can take a bit of time. You have to learn by experience. It could also improve its agent's footprint on machines. Zabbix is less appropriate if you have a huge, constantly changing infrastructure such as autoscaling. Application level monitoring is better done by other tools out there like Datadog.