① Error reporting:

zabbix_agentd [20529]: cannot create Semaphore: [28] No space left on device

zabbix_agentd [20529]: unable to create mutex for log file

After installing the agent in CentOS 6.5, it is found that the agent cannot be started. The errors in the log are as follows:

zabbix_agentd [20529]: cannot create Semaphore: [28] No space left on device

zabbix_agentd [20529]: unable to create mutex for log file

Modify/etc/sysctl.conf

Add the following line:

kernel.sem = 500         sixty-four thousand   sixty-four       two hundred and fifty-six

After modification, execute sysctl – P to make it effective.

Just restart ZABBIX agent

② Error reporting: no route to host processing

Configure ZABBIX on the client today_After agentd, register with ZABBIX automatically_On the server page, click the host list, but you find that ZBx displays red and cannot be monitored. The error message is:

no route to host

no problem is found on port 10051 of the telnet server of the client, An error is reported on the server telnet client 10050 port:
telnet 1.1.1 10050
trying 1.1.1.1…
telnet: connect to address 120.27.241.253: no route to host

it was originally lost by the client’s firewall file. Close the client’s firewall or configure the corresponding rules

③ Installation error:

Error Downloading Packages:
  zabbix-agent-1.8.22-1.el6.x86_64: Insufficient space in download directory /var/cache/yum/x86_64/6/epel/packages
    * free   0  
    * needed 131 k
  zabbix-1.8.22-1.el6.x86_64: Insufficient space in download directory /var/cache/yum/x86_64/6/epel/packages
    * free   0  
    * needed 93 k

Reason:

Insufficient disk space, failed

If the DF – LH command finds that the disk is full, delete some logs

④ Zookeeper does not plot

View log/var/log/ZABBIX/ZABBIX_Agentd.log, a large number of errors

1404:20161225:183259.913 active check configuration update from [1.1.1.1:10051] started to fail (ZBX_TCP_READ() timed out)

It was ZABBIX_The sender needs to actively send data to the server, but the 10051 port on the ZABBIX server side is blocked by the firewall. The problem of re releasing the port is solved

ZBx appears in ZABBIX custom key_ NOTSUPPORTED: Unsupported item key.

Category: Linux

2016-04-26 16:03:02

I haven’t done much ZABBIX before. In recent days, I have to monitor the master and slave of MySQL according to
basic rules http://www.linuxidc.com/Linux/2012-10/72552.htm
This can be done, but the client is ready. After restarting the service, the server can’t get the key, and the prompt is ZBx_ Notsupported: Unsupported item key.

all kinds of queries, close SELinux, release port through firewall, telnet client 10050 is connected, change agentd. After the configuration of

allowroot = 1
enable remote commands = 1
unsafeuserparameters = 1
, the service can not be restarted.
A little confused….

then it is found that the process started by the client is not listening to port 10050, so it can start the service directly by pkill – F ZABBIX
this time…

the article with the link address is below

Recently, we built a ZABBIX monitoring system in our company. Now we need to use ZABBIX to monitor the master-slave synchronization of MySQL

Now let’s talk about the detailed steps of configuration

1. First, assign a monitoring account to MySQL

mysql&> grant replication client on *.* to ‘zabbix’@’localhost’;

2. Write a script

Let’s carry out an order first

mysql -u zabbix -e ‘show slave status\G’

We choose from the output information

Slave_ IO_ Running: Yes

Slave_ SQL_ Running: Yes

These two are monitored

I tested it. When the operation data is abnormal, slave_ SQL_ Running will become No

When the slave stop is executed, both will become No

The script is as follows

#!/bin/bash
/usr/local/mysql/bin/mysql -uzabbix -e ‘show slave status\G’ |grep -E “Slave_ IO_ Running|Slave_ SQL_ Running”|awk ‘{print $2}’|grep -c Yes

I just wrote an order, and I’ll explain it

First, use ZABBIX to get all the states of the slave, then grep the two states, and then output the second column. Finally, check several yes states

Normally, there are two yes states

In fact, you will find that there is only one command, but I have written it into a script. Here I say, because if you write the command into the configuration file, you have to restart the agent after each modification, so it is more convenient to debug and modify it in the script. This should be a good habit.

3. Add monitoring items in agent

Add the following line at the end of the file

UserParameter= mysql.replication ,/home/zabbix/mysql- replication.sh

In the string after the equal sign, the comma is preceded by the key, followed by the script to be executed

After adding, restart the agent program

4. Add monitoring items on server

Go to the ZABBIX installation directory, and mine is/usr/local/ZABBIX/bin

Implementation./ZABBIX_ get -s 192.168.177.185 -k ” mysql.replication “

192.168.177.185 here is the IP address of my agent

If the master-slave copy is normal, 2 will be returned (indicating that both states are yes)

Now the server can get the status from the agent.

Now add monitoring items in the management interface

System configuration – Host – select the monitoring items of the host to be monitored

In the upper right corner, select Create item

Note that the key here is the key filled in the agent configuration file, and the data type should also be numeric

The data update interval can be set according to your own needs. After creation, you can save it

Create trigger again

Enter the trigger and click create trigger

You can fill in the name here. The warning degree is disaster

Then there is the expression. After clicking the hyperlink, select the “Edit” button

In the pop-up connection, click search after item and select the monitoring item just created

There are some rules in function. We choose last value

We set n to 2, that is, if it is less than 2 yes, the alarm will be given, and the rest will be ignored.

After setting, don’t forget to click the “add” button.

Finally, click save.

It’s configured here. At this time

At this time, we stop the replication process of MySQL, wait a moment, open the monitoring interface of ZABBIX, and select “status statistics” and “dashboard”

You can see the alarm.

We can also set the SMS alarm again, which will not be detailed here.