What needs improvement with Zabbix?

For me, Zabbix is very straightforward. I cannot think of any improvements needed. It's a very mature product. The only issue I can note is that it's Linux-based, and Linux documentation is not the best.

I'm using the free version of Zabbix, and I'd like to see more customization options, especially for setting trigger thresholds.

To improve Zabbix, adding more features to support the monitoring of modern workloads like containers would be beneficial. Many environments are shifting away from traditional setups to remote and customized infrastructures. Enhancing Zabbix to accommodate these changes would make it more appealing and user-friendly. This would help organizations like mine in selecting the most suitable monitoring solution.

The reporting features need improvement, especially detailed inventory reporting. Since it's freeware, reporting may not be a major focus. When an audit happens, we rely on the API to support the auditor. If we had stronger reporting built-in, it would provide more information for the audit process. I think that would be a good addition to Zabbix. Zabbix could use improvements in its database handling. I'm not sure if it has performance limitations in large-scale environments. We've faced performance issues when using it in a big financial institution in Thailand. The performance issues were a big problem.

In the next release, I'm hoping for features targeted towards larger users with more customizable options. Despite this, I think pre-canned reports that can be used straight out of the box would be beneficial rather than having to configure each report individually. Additionally, a deeper dive into software configurations on the machines would be useful, although I understand there may be challenges in implementing this due to scripting requirements. More documentation would also be appreciated.

There are areas of improvement. The database grows really fast. So, when you install Zabbix, you have to deal with some issues, like the database. We become pretty big very fast. This leads to optimization challenges, requiring manual deletion of network components like SNMP. We should work with SNMP. It's not as simple as configuring the host incident. Configuration can also be difficult. The use of a workflow in the configuration of items or triggers makes it difficult to use an item in a mature way. It's hard to set up a dynamic workflow. While workflows exist, setting them up for specific data sources like network items can be a hurdle. You may need to find alternative methods, which isn't ideal. For some customers, there are some personal customization challenges. But, for me, it's okay.

The solution needs to add remote features.

The integration of the product is not so easy, especially when it comes to the application database. Finding customized agents for each product needs some deep understanding. They also don't have a license for the product, but you can purchase a support contract they provide. Unlike NETSCOUT, Zabbix lacks the capability to work purely on packets, drill down on root causes and monitor any infrastructure regardless of the vendor.

Zabbix is not easy to configure, and upgrading is also an issue. But for PRTG devices, we can configure it on our own within half an hour. Besides, Zabbix is not easy to understand. The interface we need to check and what steps are to be taken to do so needs clarification. The new joiners need help learning to utilize it as they need to check the steps individually. Also, there were graphs in the earlier version, but the present version has blocks and no graphs.

I would like for this solution to be more cloud-friendly.

There is a bit of a learning curve during installation. Support could be better.

We would like to see the addition of automatic push functionality to this product. This would save time when monitoring our servers and networks as, at present, we have to manually install the Zabbix agent on any hardware to be monitored. We would also like inventory management functionality to be added to this solution.

Zabbix is powerful, but it is difficult to understand initially. There are many things that can be improved, but we might not be using Zabbix to its fullest extent. The software has more features than we need. For example, Zabbix can be used for a company monitoring solutions by many vendors. We have only one site and one company, so we can't improve all the possibilities of the software. Zabbix could add some more cybersecurity features. It isn't the same thing because network monitoring isn't the same as security. Still, it could include something like a VPN or firewall with the solution. It's a free software tool traditionally used for monitoring.

The solution needs to add features for finding loopholes or problems and their root causes.

The networking monitor is not too easy to work with. It's like switch monitoring. It's okay, yet not perfect. It's something they could improve on. I'm still using SmokePing since the network monitoring isn't so great. The initial setup isn't exactly easy.

The service module started with version six. There is room for improvement, due to the fact that this first step looks nice, yet it can be improved here for a better combination with the services and the dashboard. The event correlation could be better. From day one, we're using the API to create tickets for our operations, however, the correlation is not good, not there. A lot of the time, in that case, we are opening a few tickets for one problem, and then the correlation functionality is very, very limited. The reporting is lacking. It is there, however, it's not a real reporting from my perspective. Reporting in Zabbix means you create a dashboard and then you can schedule that and you get it as a PDF. It's probably not the reporting most operations need especially if you need more data for importing Excel or whatever. Reporting is even, in a lot of ways, missing. In the other end, we are using the API to do that now. That said, the issue there is you need the developer to do that if you decide to go that route.

In terms of user-friendliness, large maps could be more interactive. We should be able to click on some areas and move some objects. It would make it simpler to see things while analyzing some dedicated parameters.

The System Center Operations Manager can be improved. Cloud monitoring can be a bit easier, and the Zabbix dashboard can be developed a little bit more. There could be integration with Ansible Tower or AWX CLI.

As far as improvements, sometimes I get a bit frustrated when I move from a previous version to a new one because some configuration has changed—I need to investigate the documentation to deal with some configuration. But it doesn't take much time, so it's okay. I don't have any requests for additional features.

An area for improvement would be the ease of doing aggregation from the value or different devices.

Zabbix isn't a great tool for cloud-specific monitoring - its connection to public clouds needs to be improved. Other areas for improvement would be the lack of dashboards and integrations.

They need to improve the APM solution, the Application Management solution. There's a small module of APM, however, it is not an enhanced version. People usually ask for a full-fledged APM solution. The initial setup could be a bit simpler.

Zabbix does not draw automatic mapping of the network, this is something they should add in the future. There is a lot of effort that is involved in tailoring some of the settings which could be made easier.

I don't like to use the solution for the most part. I get too many false positives. I get a problem one minute and then it's okay the next. That happens a lot. I hate that. The user interface could be better. The performance needs improvement.

It's open-source software, and due to that, they really don't have legacy service monitorings like APM, or build-on capabilities, and the pure part of the transaction-related data. It is good for internal network monitoring, however, it's not for the service we are monitoring, microservices. The stability could be better. Basically, they need to provide automated monitoring, synthetic monitoring, and then APM monitoring as well as more on microservices or technology space, maybe like Java, .NET, Datadog, et cetera - these kinds of add-on instrumentations. They need to work somewhat on the dashboard and alerting side. Dashboards are not that good. They can improve on them.

My main intention is to support or use Zabbix to report patching for servers, or to somehow maintain the patching from Zabbix. Outside of the normal standard monitoring, I would like to extend patching, importing patching, and supporting patching for Windows Servers.

The performance reporting could be improved.

When using this solution in enterprise monitoring, you are able to see that there are some issues with equipment that could be causing a problem. Sometimes you want to make a root command that you do not want to be executed automatically. What we have tried to do is open an SSH session directly from the solution's interface but it is not possible. I have heard from others that there are scripts that can manage this task easily but I have not figured it out. In the area of the solution where you write the command, it is not very user-friendly and could be improved. We want the ability to write commands and execute them manually. Additionally, I would like the ability to be able to have network mapping for this solution. I use network mapping to monitor networks and there are not very many tools that are able to do it or if there is a tool, it is not easy. The color scheme of the interface could improve by allowing more variations or user customization.

Its UI needs to be improved a little bit more so that an end-user is also able to handle it. I can handle it, but others should also be able to handle it in a better way. It becomes complex when we are growing and need to add proxies. We need more scalability features and documentation for different use cases. A lot of articles are available, but they need to be in proper documentation. For example, when you have thousands of servers that have to be monitored in different regions of the world, there should be some kind of documentation to describe how you can create proxies and add them. Sometimes, when you are using the database, it can get overloaded. When the network is growing, the number of transactions becomes very high, and the database gets overloaded. There should be information about how to reduce the load on the MySQL database, which is what Zabbix is using. The market is growing a lot, and it should be enhanced for a lot more things. We are currently bringing enhancements at our end for different use cases. For example, when dockerization is going on, how can we check the logs inside the Dockers. We should also be able to monitor and check the number of logins and add features such as SSO login and two-factor authentication as a protocol. These are the security features and concerns that we have to deal with. Currently, we are developing modules to add features to Zabbix, but they should also work on these features.

Zabbix claims that there is an auto-discovery process but my team member was facing difficulty and was told that it's not really automatic, and there are some manual steps.

While there are a number of benefits to Zabbix being open source, getting that trust from the customer is a little bit on the challenging side. Implementation is always tailored to the customer and the kind of information we need from the client to carry it out can make them very uncomfortable. Sometimes the clients are not ready to share it. It can also be difficult for us to formulate a price for clients that need a solution for which there is no template. If we can implement a solution with the help of a template, we know how much time and effort it will take so it is easy to estimate a price for the customer. Without templates, estimation is a little bit challenging. In addition to the above, although there are templates available for Cisco switches, they do not work for all families of Cisco switches. These are the challenges we are facing at our company. Overall, setting up Zabbix is a challenge. However once it is setup, managing it is comparatively easy.

I think the reporting part of Zabbix can be improved in terms of more user-friendly graphics to display the collected data. Many simple users who don't know how to use Zabbix properly might get confused by the reporting, although at the same time it is very versatile for my company.

I don't recall coming across missing features. It's a fairly complete solution. The user interface could be a bit better. They could update it a bit.

Zabbix isn't very good at automation just yet. It's something we are working on. The extent of automation and the scripts that we have in ManageEngine are much more than what Zabbix gives out. For Zabbix, what we are doing is, we are creating scripts and trying to run them, however, that's not always the right way to frame it. It would be much better if there was a tool within the solution itself. The reporting is quite limited on the product. As an organization, what we did is we started utilizing a couple of connectors for the reporting. For example, we started using Power BI, and we started using Grafana. That's how we are enhancing our experience to the customers now. Zabbix doesn't give you a tool.

I am having difficulties connecting it to Grafana, as well as some of the other plugins like Kibana.

We need to set it up step by step in order to ensure that it is stable and works properly. If you want to use all of the features then you have to pay a licensing fee.

Although Zabbix is great, the documentation is a bit lacking. The documentation gets a bit messy between versions and is not too detailed, which is a bit painful for first-timers, especially when they run into issues. Zabbix depends on other open-source projects, so it would be a good idea to add more information on fine-tuning a Zabbix instance. Most of the work I had to do with limited resources is the fine-tuning of the database, OS, and other packages. Although not entirely a Zabbix issue it is a good idea to provide how-tos.

When we have a problem, we have to do a lot of research to solve it.

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.

My company wanted to do an exercise command to access IT from Cameroon. They wanted to access an FSS to a second host with second equipment that was on another coast but it is not possible on Zabbix to do it. They want to directly access from the front-end of Zabbix to access a prompt in Zabbix to an access terminal. In the front-end, there is no way to do that. That would be an important improvement.

The graphical user interface could be customized a little bit more, and also the dashboard could be more friendly. I would like to recommend the possibility of being integrated for the monitoring of the Micro Strategy System environment and clusters of machine Micro Services.

I think this solution could be more user friendly.

When our customers don't choose to go with Zabbix, it's usually been because they're looking for a solution that also contains mainstream AI, AIOps, if you wish. AIOps is a fully noted marketing buzzword, which can mean all sorts of things for different people. When I use that term, I'm talking about the requirement to analyze in-depth enterprise data. Zabbix is not an analytics platform. We've never had any issue at all integrating Zabbix with any of the existing enterprise tools, whether it be ITSM, JIRA, or an analytics tool like Splunk — it's never been an issue. Their design team does a very good job allowing people to get access to the Zabbix database and to the Zabbix Schema, which defines the data that's stored in. In the latest version, which is just about to be deployed, they are adding the capability of doing APM (Application Performance Monitoring). That's a feature that has currently been lacking — we'll have to see how it goes.

I don't know if I can remotely restart my servers or to know how a guest is doing on the server. I would like to remotely connect to the computer, and Zabbix doesn't have this capability. I have to use another software such as Emperor to log into the computer to see what is going on with the servers. I would like to see a RAS integrated with Zabbix for monitoring, also to be able to check the server and to have an automatic display to do what we need to do.

Having a more customizable interface and dashboard would be an improvement. The interface could be more user-friendly because it is can be really complicated if an end-user has to configure it. The administrator usually has to take care of that. I would like to see more SNMP and storage support. Application monitoring should be included in the future. I would like to see voice telephony monitoring and database monitoring. The reporting functionality is limited.

The NetFlow integration really could be improved upon. In general, integration with other solutions and services needs to be worked on. The documentation could be improved. I find that it's a bit limited. It runs on a Linux backend, and we're using it to monitor our Windows platforms. For people who are not familiar with Linux, there may be issues. You do need to have a bit of Linux knowledge before beginning, otherwise, you may have problems working with it.

In the near future, we are going to see integration with its cloud platform. We would like to use it for RBI, for service and credibility reports for our manager and management staff, and we also hope to use it to deliver customer scripts, etc. Currently, because we monitor over 5,000 hosts and servers, we need to use open source.

Currently, we're looking for more AI and ML related capabilities, which is missing from Zabbix at this point in time. There are tools out there with plenty of AI and ML capabilities and this is something that Zabbix needs to come out with. That would make this a great solution to sell. Essentially, they are trying to achieve automation by introducing external RP bots. If Zabbix could fix this issue, it would be a fabulous solution for us to sell and for companies to use.

If Zabbix had a better dashboard then it would be nice.

The server monitoring could be better. I'm not well versed in Zabbix, so it's difficult for someone like me to handle the setup. They should work to make the implementation much easier. It would improve the useability of the solution. The solution needs to have a higher quality of performance. Performance is very important to our clients, and this solution doesn't perform very well. The quality and code are difficult. The solution does not scale well.

The main problem with Zabbix is that you have to spend time writing templates for all of the products that you have. It would work great if I had the time to do it, but considering that I have all kinds of different devices, I don't want to spend the time doing it. If they included more prebuilt templates of devices that are in use today, then it would be very helpful. For example, they have a template for a SonicWall 3600, but they don't have one for a SonicWall Tz300. They do have some base templates but they are not customized for these specific products. For some products, I was able to download one, but not for all of them. The documentation needs to be improved. The deployment is complex and should be simpler to complete.

I would like to see a more flexible mobile client, and better HA out of the box. It would also be very interesting to have a GUI on the proxy server[s] and the ability to have the zabbix proxies send it's data to multiple zabbix servers. This is very valuable when a site is completely disconnected from the wan.

There are a lot of things that can use improvement which is why we are seeking a new solution. Network monitoring is a problem. It gives too many false positives. For example, it notifies us that a server is down while I'm using that server — the server it claims is down — to do the search. A moment after the search is complete, everything is OK again. It's called flapping. It has some flapping control, but it's not as good as other products. I used to use Icinga and it is a better product in that respect.

Open an SSH session from the web interface.

It's OK for me as is, but some frontend improvements would be nice.

Multitenancy.

It is a little complex to config and implement so I'd like for them to make it easier to config, for example, a configuration for SNMPv3 or trap listener on it.

I want Zabbix to improve the UX/UI. Zabbix doesn't use a JavaScript chart for images, and I want them to improve this.

I wish it had more options for personalizing the product, such as being able to include your company's logo without editing text files. More templates for specific devices.