Zabbix Agent active mode monitoring

Zabbix Agent active mode monitoring

Zabbix_server collects data from the Server when there are too many hosts. Zabbix has serious performance problems, mainly as follows:

1. When the monitored end reaches an order of magnitude, the Web operation is very slow, and 502 is easy to appear.

2. layer break

3. Too many processes (Pollar) are enabled. Even if the number of items is reduced, adding a certain number of machines will also cause problems.

Therefore, we will consider the following two optimization aspects:

1. Add a Proxy Node or Node mode for distributed monitoring

2. Adjust Agentd to Active Mode

Because the first solution requires adding physical machines, first try the second solution.

1. Configuration Adjustment of zabbix_Agentd.conf on the monitored end:

1. LogFile =/tmp/zabbix_agentd.log
2. StartAgents = 0 # client agent mode, which is only in active mode. When the value is 0, the monitored zabbix_agentd does not listen to the local port, so the zabbix_agentd process cannot be viewed in netstat-tunpl.
3. ServerActive = **. ** # ip address of zabbix_server
4. Hostname = test_host # Important: Host Name
5. RefreshActiveChecks = 1800 # cycle for the monitored end to obtain monitoring metrics from the server
6. BufferSize = 200 # size of the storage space of monitoring information on the monitored end
7. Timeout = 10 # Timeout

The important parameters are ServerActive and Hostname. ServerActive specifies where the data collected by Agentd is sent. Hostname must be the same as the host name when the Server is added, in this way, the Server receives the data to find the corresponding relationship.

Ii. Server zabbix_server.conf Configuration Adjustment:

StartPollers = 100

First, reduce the process of actively collecting data. The original process is more than 700.

Starttrapi pers = 200

Then, you can increase the number of processes that process the data pushed by Agentd.

3. Adjust the Template

We recommend that you clone a Template OS Linux Template to modify it. The adjustments are as follows:

Then modify the cloned Template Name.

Then modify the monitoring mode of each item.

Select All

Then select Mass update for batch Modification

Modify here

Change to active mode, save

Then select the automatic discovery Policy

Change the monitoring items to active mode and save them.

Click items

Change the monitoring items to active mode and save them.

This template is ready.

4. Add an active host

Create a new host

Enter the hostname in zabbix_agentd.conf of the monitored end as the host name, agent type, IP address as 0.0.0.0, and port as 0 for saving.

Page 2 associated monitoring template

Save

Finally, start the zabbix_agentd service on the monitored end.

Check zabbix_server.log to check whether the host can be monitored normally.

If the monitoring fails, use tcpdump to capture the data packets on the monitored end. If the data packets cannot be captured, check the firewall configuration.

You can also view latest data to check whether the monitoring value sent by the monitored end can be obtained normally.

In the host list, it seems that the monitoring type of the monitored host cannot be displayed. However, monitoring is not affected.

Now, the test is complete.

Some Zabbix Tutorials:

Compile and install Zabbix2.4.5 source code in Ubuntu 14.04

Install and deploy the distributed monitoring system Zabbix 2.06

Install and deploy the distributed monitoring system Zabbix 2.06

Install and deploy Zabbix in CentOS 6.3

Zabbix distributed monitoring system practice

Under CentOS 6.3, Zabbix monitors apache server-status

Monitoring MySQL database Parameters Using Zabbix in CentOS 6.3

Install Zabbix 2.0.6 in 64-bit CentOS 6.2

ZABBIX details: click here
ZABBIX: click here

This article permanently updates the link address: