What is our primary use case?
I'm currently dealing with three implementations - two are on cloud and one is on-premise. Our clients use the solution to monitor IT infrastructures, IT networks, applications, and cloud as well as containers.
I have a very exciting additional use case as well that we're working on, using Zabbix 5.x for internet of things (IOT) monitoring. It supports MQTT and ModBus, which allows us to monitor IOT devices.
The flexibility of taking monitoring data from such a wide variety of platforms - beyond traditional IT - makes Zabbix a highly flexible solution.
I'm a consultant of this solution as well as a user. I'm a Zabbix certified professional and director of a Zabbix authorized reseller (Nobius) in the UK.
How has it helped my organization?
As our business has grown, we use Zabbix to monitor our Zabbix implementations - a 'manager of managers' setup. This has given us better resource utilisation and improved our service delivery.
What is most valuable?
The ability to take data from multiple sources. The Zabbix agent is probably the most lightweight monitoring agent available. The agent itself is extensible providing simple expansion capabilities to support new use cases. Alongside the built-in agentless monitoring via SNMP, SSH, WMI and others, this means we have a solution that has no limits as to the data that can be ingested and alerted on.
What needs improvement?
If anything could be improved, it would be some of the terminology that is used in the documentation. The documentation is good, but it's been translated into English and occasionally suffers from terminology issues. There are additional features that the commercial software has that Zabbix doesn't. Full AIOps isn't cheap, things like machine learning and artificial intelligence attract a massive price premium and are rarely implemented properly. But they are major, major features.
For how long have I used the solution?
I've been working with Zabbix for over four years following 25yrs as an IT Monitoring specialist within user organisations, consultancies and major global software vendors.
What do I think about the stability of the solution?
The stability is unparalleled. Processes are all well-behaved and logging is clear and succinct. Support for separating the front end, server and database allow resources to be load-balanced and clustered.
Zabbix runs equally well in cloud- and on-premise configurations.
What do I think about the scalability of the solution?
The solution can take hundreds and thousands of values per second, so scalability is excellent. The ability to add proxy servers to distribute the data handling load is impressive and they are very straightforward to set up. This also adds to reliability in distributed environments.
How are customer service and technical support?
If you consider that Zabbix is open source, their technical support is fantastic because the people, the developers of the application, are the guys who do the technical support as well. You don't have to wait to get through different levels of technical support. You get a very, very knowledgeable person on the phone straight away which is a big plus.
Which solution did I use previously and why did I switch?
Yes, previously used and consulted on HPE SiteScope, HPE Operations Manager, Micro Focus Operations Bridge and others.
Sitescope is useful, but doesn't have the muli-tenancy or proxy facilities and is agentless only making it unjustifiable in an MSP environment.
Operations Manager was too complex to maintain and became obsolete.
Operations Bridge is resource hungry, complex to install and configure and extremely expensive.
How was the initial setup?
I've deployed so many times that the initial setup is straightforward, but I would say that for someone who is totally inexperienced in Linux, it can be a little time consuming. If you understand a little about Linux, then it's no problem. A full system can easily be configured in two hours but it took two days the first time I did it. If you're not a technical person you can still install it but it will likely take some time.
As an example, configuring SNMP trapping into Zabbix needs configuration outside of Zabbix itself. This is not complex, but can slow down the process for inexperienced installers.
What's my experience with pricing, setup cost, and licensing?
The software itself is open source, it can be easily downloaded, and use it with no limitations.
Be very careful about using the "appliance" configuration in a production environment. It is only suitable for evaluation or very small environments.
Invest in support, training and consultancy from Zabbix or from third parties. Architecting a robust, resilient and secure monitoring platform from day 1 will save time and money at a later stage.
Zabbix and 3rd parties offer far more than a traditional support contract. No other organisation in my experience includes pro-active and on-site support as a core part of their offerings.
Which other solutions did I evaluate?
Nagios - too much development effort to maintain and configure.
What other advice do I have?
I believe it's crucial to plan the implementation. Just because the software is free, you shouldn't just install it and let it run. Plan your implementation carefully and you'll get more out of it than you ever thought possible
I would rate this solution a nine out of 10.