Enterprise monitoring in All for One

Monitoring of IT systems, including SAP systems, databases, operating systems and physical hosts is one of the services offered by All for One. Monitoring is a regular checking process aimed at keeping systems and services in the best possible condition. This helps minimize the likelihood of unplanned interruptions in the operation of critical services. Historical data also helps optimize system performance or diagnose potential failures.

Detecting abnormalities in advance makes it possible to address issues before they arise. In emergency situations, the monitoring system is an additional tool for analyzing the situation to track down problems before they escalate too much.

When determining the scope of monitoring, we need to pay attention to many factors, such as: how much time it takes for the server to respond, how long it takes to send queries to databases. Then, the main elements for monitoring should be identified. Their choice depends on the business services hosted on the server and the selected system architecture. This work requires meticulousness since inaccurate monitoring gives a false sense of guarantee of service availability, but also weakens our vigilance when problems arise.

Nagios software, which we used for many years, is one of the most recognizable systems in the industry. It is an open source tool, which makes it very flexible because we can adapt each monitoring element to our needs. Nagios offers many pre-built scripts for monitoring multiple elements of the infrastructure, but the biggest advantage is that they are customizable to our needs and those of our customers. Each message, threshold or command can be customized so that the monitoring meets all expectations and ensures trouble-free operation of the system.

Verification of systems by Nagios monitoring can be done using an agent installed on the machine or without it, in passive mode. It detects failures of networks, protocols, malfunctioning services, processes, and background tasks. With properly written scripts, it enables monitoring of databases, SAP services, operating system metrics and all the processes it performs. Monitoring also includes the status of servers, log analysis, operation of applications and virtual environments. The actual configuration is done using text files and the dashboard is used for reporting.

Due to the need to maintain a higher level of certification and compliance with ISO policy, we decided to migrate monitoring to Zabbix. The first version was released in 2001. Its developers are driven by the belief that monitoring should be powerful and accessible to everyone. Zabbix is a flexible tool for monitoring IT infrastructure. It is an enterprise-class solution used by the largest companies in the world.

Zabbix is a highly scalable system that can support both small businesses and large corporations with extensive infrastructures. It is used to collect, analyze and visualize data. The monitoring is based on templates that we can customize globally, addressing a deficiency we experienced in Nagios, where configuration had to be adjusted per host in text files. This required a significant amount of work during the installation or migration of the new customer’s systems as part of hosting at the All for One Data Center.

The solution provides a unified view of the entire infrastructure. Zabbix allows the use of a variety of templates, i.e. sets of pre-determined monitoring metrics, and additional elements can be detected automatically through a discovery process. They automatically identify existing databases and configurations after entering basic data for connection. The system offers a wide range of indicators that can be monitored. Currently, we can set up full monitoring of the new system or even the entire landscape in just a few minutes.

Why Zabbix? There are at least a few reasons, one of which is the fact that the solution is open source. It is a stable software that has been operating on the market for years, with a large community working towards innovation and stability. One of the key advantages is a very rich technical documentation and a website, which was created by the developers. Users can share their infrastructure monitoring methods in the form of ready-made templates and scripts. There are also forums that provide a lot of knowledge and solutions, which can be very useful when implementing configurations. The activity of a large community of Zabbix users translates into continuous development of the system, which means that we gain access to new features and solutions. A strong point is the ability to run custom scripts.

The solution has a variety of ways to present data, such as maps, charts and reports, which help in data analysis. With notification and alert mechanisms, Zabbix allows for quick response to problems that arise, which minimizes downtime and increases system performance.

Migration to Zabbix is a large project divided into smaller milestones. The first one was to install and implement the system. The task required a scalable and efficient architecture. After the installation, we proceeded to the process of tuning all components of the Zabbix system to obtain a solution that was efficient and tailored to our needs. In the same step, we prepared procedures and standards that proved to be extremely useful in the process of migrating the monitored architecture from Nagios. We created basic network rules, and verified whether communication with customer systems in our data center and outside it would be possible for all systems. Furthermore, we established monitoring procedures and standards.

In the next stage, we proceeded to install Zabbix agents on end systems, along with the necessary configuration. At the same time, we decided to set up a standard of tailored monitoring, creating most templates by ourselves. After building templates monitoring operating systems, databases, SAP systems and other services, we were ready for the next stage: migration between systems. This process continues, gradually moving elements from Nagios to Zabbix. In parallel, as experience is gathered, monitoring is being refined and enhanced.

Currently, Zabbix is fully integrated with our ticketing system. Working with the monitoring system has gained a new quality. When an alert is triggered, Zabbix verifies the alert-related system and classifies the alert based on the following variables: priority, team, category. The ticket is then transferred to the ServiceDesk and assigned to the appropriate team. When the alert stops, Zabbix will update the ticket, adding the necessary information. The ticketing system also allows for identifying recurring issues and responding appropriately by increasing resources or verifying the correctness of system operation.

The process of monitoring IT infrastructure has always been a challenge that requires careful planning and adaptation to a dynamically changing environment and needs. We offer a wide range of monitoring services to All for One Poland customers, and the migration from Nagios to Zabbix allowed us to increase the flexibility and efficiency of our operations. The implementation of Zabbix has been a gradual process and required a considerable amount of work, but we are now able to monitor our customers’ infrastructure more effectively than ever before. Our monitoring is becoming more efficient and reliable.