In every company you will find a wide range of different types of servers, such as web servers, databases, file servers or mail servers. Often they perform business-critical tasks, and are essential for the success of your business. If IT managers do not take various factors into account when monitoring servers, they will be taking an unnecessary risk. To ensure the smooth running of IT processes, the servers use case should be included in the IT monitoring.
In addition to the intended use of a server, it also makes sense to consider the underlying technology, for example, whether it is a physical platform or a virtual machine – on-premises or in the cloud. Simply monitoring the basic performance data of a server, such as CPU, RAM, memory usage and network connections is therefore often not sufficient. Depending on the type of server, it may also be appropriate to include information about applications, services, and processes in the monitoring. In addition, interactions and environment variables often play a role and can influence the function of the server. These include, for example, network components, storage, or power supply.
Checkmk is easy to configure but also really flexible for all kinds of special requirements.
On this page you will learn which factors are really important for monitoring, and how you can best combine these to set up an optimal observation of your different server types. Practical experience has shown that this procedure makes perfect sense. However, with the wrong IT monitoring tool the workload can increase quickly, especially if the tool does not support the users.
Checkmk offers an easy to use user interface, and allows, among other things, the administration of systems via host and service groups, or with folder structures. Thus a large number of systems can be efficiently managed in monitoring. In addition, Checkmk is able to query a wide range of information. The Checkmk RAW Edition is completely free and open source. The Checkmk Enterprise Edition is even more responsive to the needs of professional users, and can be precisely adapted to different types of servers.
Key points when monitoring different types of servers
Check the performance of your web servers
Make sure your website is always available and provides a rich user experience.
Assure the continuous flow of information and avoid security risks.
File server monitoring
Protect your company assets against unwanted access and make sure your employees can access the data they need.
Monitoring mail servers
Reach full communication capabilities and assure the function protection mechanisms for sensitive information.
Basics of web server monitoring
Web servers are an example of both the dynamics of digitalization and the challenges of monitoring. On the one hand, after 30 years the World Wide Web still relies on the Hypertext Transfer Protocol (HTTP), on the other hand the possibilities of the Internet have changed significantly. Almost every product or service is now sold online. Web servers or cloud products with similar functions are central to the provision of digital products.
Websites, web services and web applications are the most important elements for the success of an organization. Especially if you are delivering content to end customers, performance is the measure of all things. Poor availability quickly leads to user disaffection, because in our always-online world, an alternative offer is often just a mouse click away.
To ensure that content and services are available at peak times, you should keep an eye on your web servers properly. Compared to other types of servers, response times are particularly important here. You can use monitoring to ensure that loads are optimally distributed. In addition to the web system, you can also monitor load balancers. By combining the observation of web servers and load balancing in one tool, you can identify potential sources of danger at any time.
Checkmk captures the right indicators and helps you to set targeted alarms to ensure the optimal functioning of your web servers. Plug-ins for all common programs, such as Nginx, Apache HTML or Microsoft IIS are available. These give you an optimal information basis for your monitoring. Also read our blog 'how to monitor your web server'.
Keep an eye on these web server metrics
Like on any other server you should check CPU, RAM, disk space and network data for your web servers. However, metrics related to connection requests play a special role. Long response times, for example, are a bad sign, and indicate problems with the server or homepage design.
In addition, large data packets should be avoided in the response size. Especially with apps on mobile devices, large amounts of data in combination with slow Internet connections are a nightmare for users, but are difficult to detect during development and testing.
You should also keep an eye on the number of active connections on a server. By monitoring incoming and outgoing traffic, and access to a particular server, you can take important security precautions.
In addition to unwanted requests, companies can see if they need to adjust their resources before unnecessary slowdowns or unavailability occur. Monitoring, therefore, helps to ensure good performance.
The architecture of Checkmk allows the setting up of an efficient monitoring system in a few minutes, without any unnecessary added load on the monitored systems. Checkmk obtains its information via its own agents or via SNMP. The advantage of the Checkmk agents is their ability to retrieve information for individual programs, tasks and services. In addition, plug-ins for web server solutions are also available.
Checking response times and the certificates of web servers
Companies should anchor the checking of the availability of their web servers in their IT monitoring system. In this way they always have an eye on whether their servers deliver the desired response times and send the expected content in response to requests. If something goes wrong here, IT teams receive immediate notification.
In addition to response time and content check, IT teams can use the active HTTP check to verify the validity of SSL certificates on servers. This prevents customer complaints, especially in the areas of e-commerce or banking, because expired certificates on web servers lead to either a security warning or an error message in the users browser when the website is called up. Both should be avoided in any case.
Checkmk can easily observe HTTP connections and URLs. The active check needs no agent on the server and runs with any web server – such as Apache, Nginx or Lighttpd. The HTTP check provides a variety of ways to check the URLs.
The insights you are getting from it are great and help to resolve problems even before they are real problems.
Analyze data with your web server monitoring tool
Especially with web servers, it makes sense to keep an eye on the temporal development of accesses and data streams. This is the only way to properly check the performance of your servers. This also includes the exchange of knowledge. Anomalies, poor availability and other problems are not always automatically exchanged between IT teams and reported to the responsible department. A disadvantageous configuration can often only be detected in context.
The mere collection of data does not provide any insight into the operation of your own servers. Although the number of connections, active users and sessions, and many other data are suitable for better planning of your own infrastructure, you still need to visualize the metrics correctly, otherwise no evaluation is possible. Only then will a web server monitoring tool be suitable for evaluating and improving server performance.
Checkmk records measured values and provides this data in meaningful graphs. Their use is easy and intuitive even for users without monitoring experience. The analysis is easy thanks to its integrated graphing tools, and enables the quick detection of deviations. If required, the information can also be forwarded to external tools such as Grafana. In Checkmk you can also combine and customize graphs – for example to investigate temporal interactions. You can see more details in our video tutorials.
Basics for the monitoring of databases
For a long time, a database server was simply a piece of hardware that developers and administrators dismissed as a minor matter – so their monitoring was limited to availability, storage space and rather general metrics. The increasing use of IoT and Big Data applications, as well as mobile applications, means that databases must, however, meet the highest demands for performance and scalability.
To keep the limitations of the CAP theorem as small as possible, database monitoring is essential. Accordingly, data availability in distributed systems (consistency), response times (availability), and the number of network nodes (partition tolerance) must be part of the monitoring.
In addition to ensuring availability, businesses should constantly optimize the flow of information. More and more players need access to data, which means that the number of queries is also growing. Database system administrators rebuild or adapt databases more frequently. They are also under time pressure, and need a monitoring tool that helps them to tackle their challenges with a few simple steps.
Checkmk is perfectly suited as database monitoring tool. In addition to details on memory, buffer cache sizes and connection times, Checkmk allows accurate insights both over short periods of time and for long-term investigations. In addition, Checkmk supports structural databases such as Oracle, as well as NoSQL approaches such as MongoDB or Couchbase. The number of official Checkmk plug-ins is also constantly being expanded.
Monitors eveything from network equipment, Power supply, Hardware, OS, virtualisation technology, databases, storage systems, backup and even modern stacks like docker. Also helps with automated SLA reports und software/hardware inventory.
Monitoring of heterogeneous database landscapes
Most organizations use several types of databases in parallel. Familiar examples are Oracle, Microsoft SQL, Sybase, IBM DB2, SAP HANA, or PostgreSQL. In addition, among SQL databases there are also open source products such as MariaDB or MySQL. There are also non-structural (NoSQL) databases such as MongoDB or Couchbase.
For database monitoring, performance monitoring is crucial for almost all types and manufacturers. Depending on the technology, different database attributes are relevant for this. With Checkmk you get all necessary data and are flexible in the database selection. You can transfer the monitoring information itself to Checkmk from external sources (via network ports), or from internal ones (via the operating system).
Checkmk provides standard checks that are already implemented in the check plug-ins. No database know-how is required for monitoring. See what is currently running on each of your database instances, and visualize the desired processes in a user-defined dashboard. For example, developers and DevOps engineers can use this to identify early on whether they need to take certain requirements of the production environment into account during software development.
Database monitoring in practice: Oracle
With structured databases – such as Oracle databases – Checkmk helps administrators quickly and reliably monitor performance and availability, and provides information about user statistics. If errors occur, or if a defined database size is exceeded, the software notifies the responsible administrators via alarm and – depending on the configuration – can also perform corrective actions.
Checkmk has more than 20 plug-ins for Oracle alone, and it can retrieve and transfer information, for example, about user activities, tablespaces (usage, status, size and growth), ASM disk groups, sessions (user sessions, wait times, CPU usage of certain users, number of memory slots and tablescans etc.), and much more into the monitoring.
Like many other software suppliers, Oracle provides its own tools for monitoring, but these are limited to its own databases. The maintenance of several tools leads to a considerable additional administration effort. Furthermore, no context data can be included. With Checkmk you have an uniform information base which you can adapt for different purposes. In this way, you can recognize possible interactions with applications, other servers or network components.
Monitor changes and database migration
Since databases are so strongly in the focus of digitalization, there are numerous innovations in this area. The requirements are constantly growing, which leads to companies continuously evaluating their databases and adapting new approaches. Cloud and container approaches, as well as new approaches from the open source community, are pointing the way forward.
Especially non-structural databases are on the march. This is no surprise, after all, even very large amounts of data can be created and processed in NoSQL databases without any loss of performance. Very good throughput rates can be achieved with suitable hierarchies and query languages. On the other hand, it is possible to transfer legacy data records, even if they are only available in a relational format.
This makes it all the more important to monitor all databases – both before and after a changeover. Checkmk integrates with common structural databases. Integrations for NoSQL databases like MongoDB or Couchbase are also available. This allows you to adapt your monitoring seamlessly, and to analyze states before, during and after an adaptation.
Basics of file server monitoring
File servers are part of everyday life in almost every office to provide users with shared access to files. Protocols such as FTP, SFTP or WebDAV allow the exchange of information in a computer network or via the Internet. Since file servers often contain internal company data, their monitoring is particularly important for security reasons.
The focus here is on approvals and accesses. It is bad if employees do not have access to required servers, but it is usually a threat to existence if the wrong people can access company data. In monitoring, you can create alarms if an unauthorized IP address accesses servers, or if certain data is changed.
It is also useful to adjust threshold values for basic measuring ranges. Especially for storage space, warnings should possibly be followed-up earlier than with other server types, since user behavior can never be predicted exactly. Full hard disks or overloaded servers mean poor productivity because employees cannot perform their jobs effectively. Availability issues also have a negative impact on productivity.
Checkmk integrates with file server protocols such as FTP and allows accurate monitoring of connections. As a holistic monitoring solution, Checkmk not only monitors servers and networks, but also integrates, for example, with Hardware Security Modules (HSMs), to ensure the comprehensive protection of information. Especially for file servers holding sensitive information, you should rely on strong encryption throughout.
Approaches for the monitoring of file servers
The operating system is suitable as a first information resource. This is usually selected according to the IT environment and number of users. Microsoft Windows, Linux or macOS are used in practice. The exact version or distribution also depends on the available hardware and the number of users who ultimately need to access the server simultaneously.
Checkmk can transfer relevant information to the monitoring via its agents. It is also advisable to use the plug-ins mentioned earlier for protocols like FTP, SFTP, SMB or NFS. This allows, for example, automated active checking against previously-agreed time limits for download servers. If data is not available after a predefined time, an alarm is sent.
Monitoring also allows you to see which data and versions are available. In this way, you can ensure that unwanted changes or manipulations are impossible.
The simple administration via rules in Checkmk prepares the monitoring for the various functions of file servers. For example, you can schedule time windows for backups in which large data movements between two specific servers do not lead to alarms, or Checkmk sends messages if this situation does not occur.
The automatic agent distribution is saving much time and avoids using further tools like external packaging.
Monitoring of mail servers
Despite new competition from B2B messaging apps like Slack, email is still the most important means of communication for businesses. Mail servers are a central element for coordinating workflows between colleagues, and facilitating agreements with customers and service providers. Businesses depend on reliable email traffic. For example, if documents are sent shortly before important appointments, the failure of the mail server can even jeopardize the success of negotiations.
Monitoring Postfix, qmail, Exim, Microsoft Exchange and other mail transfer agents is therefore an absolute must for IT managers. In addition to availability and performance, there are important details, such as ensuring that mails have been delivered and that they are securely archived. Just as important is the monitoring of server services such as POP3, IMAP or SMTP.
Checkmk offers official plug-ins for all common mail servers and server services. This allows companies to ensure that their mail servers work efficiently, that encryption and backups work properly, and that digital correspondence is guaranteed at all times. Also, Checkmk is not dependent on notifications only by email. If desired, you can be informed about incidents 24 hours a day via slack, SMS, fax or any other method.
Analysis and adaption of mail servers
Mail servers and email traffic are changing. Since the introduction of Microsoft Office 365 at the latest, more and more companies are moving their mail servers to the cloud. At the same time, there are new requirements for the storage of personal information as a result of the General Data Protection Regulation (GDPR) and other data protection acts. Therefore companies should be able to check the performance and information flow on their mail servers.
With the integrated graphing and reporting tools of Checkmk, you can reliably detect anomalies and critical events. In addition, you can also track the effects of changes, for example, if you move mail servers to the cloud or change mail servers. Furthermore, you can always keep an eye on the validity of certificates and recognize if mail traffic is routed through unencrypted ports.
Checkmk can be set up within minutes, and it allows you to monitor all mail applications. Thanks to its easy-to-read dashboards, you have an overview of all important metrics for monitoring your mail servers, and receive targeted alerts when there is something to report – and only then.
Frequently-asked questions about adapting the monitoring for different types of servers
Is Checkmk also suitable for monitoring Java application servers?
Yes, almost all common application servers such as JBoss, Oracle WebLogic, Apache Tomcat or IBM WebSphere can be monitored with Checkmk.
Does Checkmk provides agents only for Windows and standards Linux distributions?
A wide range of agents are available for different operating systems. Systems such as macOS, AIX, Solaris, HP/UX, OpenBSD, FreeBSD, NetBSD, OpenVMS or OpenWRT are also supported. Plug-ins can also be easily extended and adapted.
Is Checkmk suitable for managed service provider?
Yes, the Checkmk Managed Service Edition is optimally suited for IT service providers and consultants. It offers advanced instance and customer management, comprehensive dashboards, and strict data separation. You can also use it as a white label solution and customize Checkmk according to your needs. Furthermore, all editions of Checkmk are designed for maximum performance, and support distributed monitoring.