Client and server monitoring with Zabbix

[Sanju C]

You should have got an idea from title, what I want to know. Even a good monitoring tool is only worth half, if you can not manage it easily. Many of my friends told me that with Zabbix, this is not a problem: The free monitoring tool you can easily use and manage via web frontend. But the problem is that I am not having much knowledge about the zabbix. I am here to get more ideas about adding clients. I am also eager to know about an importing and exporting functions. I am sure that you members will make some help for me.

[VinFanatic]

Go to the front in the browser http://<zabbix-server>/zabbix/ on and log in as an Admin user. If you have not yet changed the password, it is the first sign for zabbix. You should change course immediately. In the Administration screen, you can also change the language, if desired. For each user can customize the way the language. The buttons control, inventory, reports and application are available for everyone. However, not all the possibilities that lie under these links. Once you have registered with a particular user to expand the possibilities of the administration and configuration. The user management administration is hidden behind. Here you can as you wish to add new users. Zabbix are three types of users: Zabbix user Zabbix Admin and Super Admin user to type the users have different rights depending. Also, you can store, learn how the user of possible events is. In addition, you can choose to receive messages which weightings of the user. This can be done via e-mail, SMS or Jabber.

[Joko]

Under Administration - General, you can first decide how long tool information in the database is to store the monitoring. Furthermore, the tab configuration - system groups to look at. Here the developers have already more than 40 useful templates provided. These are useful and can save time in setting up systems to monitor a lot of time. Likewise, the programmers have included a import/export. Subsystems, you can now add to the monitored system start. It makes sense when it is ready for a Zabbix agent. You can use the client with either the DNS name or IP address to address. Likewise, you could change the port if you have switched on the client. To save time, you have the option of the client with a template link. In most of the templates contain much more than you need. But with a simple click, you can disable unused requests. It is also quite useful to map the specific computer groups. Particularly in larger environments the administrator retains a bit of confusion. It may also not hurt to provide additional system details. For example, you can store the operating system software running and the MAC address.

[joel84]

If you use the templates, you can probably disable the monitoring of standard servers rather elements. The Linux template contains, for example, the monitoring of the directories /, / home, / opt, / tmp, / usr and / var. To keep an eye on these directories, you do not create even more hands. Often some of these directories are not on their own partitions. So you could turn off monitoring if necessary. Another example CPU load, running SMTP server , free memory and swap, FTP server, Apache, network traffic monitoring. The checksums of the files / etc / passwd, / etc / services, / usr / bin / ssh and / usr / bin / sshd checks the monitoring tool is also standard. Should change something in these files, you can Zabbix can sound the alarm.

[afidelino]

The Windows Template monitored by default include the drives c: and d:, CPU usage, memory and file access speeds. If you want to monitor other services or daemons, you can add new ones. In our example we want to monitor the availability of the Samba service. This one needs to know that the process in case of an Ubuntu server (9.10) is smbd. This can vary from system to system. From the documentation we can see that it processes the key proc.num counts. The service is to be monitored in square brackets: proc.num [smbd] counts as a key, if enabled, now all samba processes running on the server instance. It makes more sense to assign user-defined elements of certain applications. This monitoring will be clearer.

[T0tal L0$$]

A trigger shall come into force when a particular event occurs. It triggers a certain action. For example, realized the Zabbix server that Samba processes jump to zero. This means that no process is running the file and print server more. Zabbix can respond to this event and triggers a pre-configured action. As already mentioned, this e-mail, SMS or Jabber happen. If the number of processes one again or higher, you can turn to be notified. Can also cause you to declare as active or inactive. The term itself is in curly brackets. The box will be the first to be monitored computer, then the key and then what to monitor. For our example, saw the expression of a newly created trigger like this:

{Zabbix server: proc.num [smbd] last (0).} <1

You can assign different triggers different priorities. In the case of a file and print server, it would certainly be a disaster if users can not access data. With the actions you can define how to deal with Zabbix a triggered event. For our example is as a source of "triggers" are used. will be activated should continue to benefit as a precondition, that the trigger "down Samba". This is supposed to send the software to the user Admin. The service should be back, the administrator would also like to have a message. Thus, we can configure the recovery message with the same. In the message box, it is useful to use meaningful lyrics. It is of value if the recipient of such mail directly know what is going on. Administrators known to be reluctant to advise on error descriptions.