Do you have Linux servers to host corporate applications or e-commerce websites? So you know how important it is that they are always available, their proper functioning has become a key point in the operational excellence of a company's IT infrastructure and systems.
Being familiar with the response time, the number of visits to a website or the demand on the server is now part of a holistic view of IT, improving the diagnosis of problems in real time.
But... How can you monitor them optimally? Surely you have heard of several possibilities, but the most important thing is that to have your Linux servers under control in WOCU you do not have to change the way you work. If you want to know what I mean, keep reading.
As we said before, one of the most important points that we must take into account when monitoring Linux servers, is to know which are the protocols that these servers have enabled, and from them, decide which one we want to use to obtain the typical monitoring metrics of health or performance of the underlying operating system.
At WOCU we have included monitoring packs that collect data from Linux servers through the following protocols or technologies:
Although the open-source NRPE agent offers great features on most platforms, in some cases it is not possible to install it. In this situation, both the Linux-SSH pack and Linux-SNMP and/or OS-Resource provide us with agentless monitoring methods.
In those infrastructures where it is possible to use the SSH protocol to carry out the monitoring of Linux servers, we will use the Linux-SSH pack, which is configured by establishing a connection without a password between the WOCU server and the remote server. To do this, first the administrator of the monitoring platform must log in to the WOCU server and create the SSH keys of the user who is going to monitor the infrastructure and then distribute the public key to the remote Linux servers. By default the Linux-SSH package uses port 22 to establish the connection, although it can be changed as needed.
As a second option to monitoring Linux servers without agents, you can use SNMP monitoring, sometimes there may be prohibitions regarding installing NRPE agents or using the SSH protocol for monitoring servers.
SNMP checks are performed only through the UDP protocol.
Additionally, if we install/enable the Net-SNMP client on the computers, we will be able to access additional information to what we obtain with the Linux-SNMP pack.
The Net-SNMP client on Linux servers provides a wide variety of performance information about the SNMP protocol. In addition, the agent can be queried for a list of installed packages on the system, a list of processes currently running on the system, or the system's network settings.
The Host Resource MIB (HOST-RESOURCE-MIB) included with Net-SNMP presents information about the current hardware and software configuration of the computer.
Each of these packs will provide us with information on different metrics, some are common to all packs, others, however, are only available in one pack or another, although additional services can normally be added to (almost) all of them.
For a better understanding of the metrics that are monitored according to the pack, we have made a summary table where you can check what each one includes:
Img.1 Comparative table of metrics by monitoring Pack
As always, you have complete information on each pack in our documentation, in the WOCU user manual.
Do you still have doubts about how to monitor Linux servers?
Contact us and tell us your needs, we will be happy to hear your suggestions :D
