CODEWITHSUNDEEP
zabbix
Andrew
Andrew

Reputation: 61

zabbix_sender: send value error: ZBX_TCP_READ() timed out

I have some problems sending my test data to Zabbix Server.

My configurations are set up, even have Zabbix agent installed and correctly working (sends data for monitoring CPU, memory,...).

This is the situation: I have a Zabbix installed on a Debian VM, and configured a host with correct IP, port, item(Zabbix trapper). I want to send a value just for testing from my Windows 10 PC using "zabbix_sender"; later I want to find a way to get data from a .txt file for monitoring.

Used command from my cmd:

zabbix_sender -vv -z XXX.XXX.X.X -p XXXX -s "IT-CONS-PC4" -k trap -o "test"

Error:

zabbix_sender [8688]: DEBUG: send value error: ZBX_TCP_READ() timed out

Did someone else had this issue?

Upvotes: 3

Views: 30317

Answers (3)

Juan-Kabbali
Juan-Kabbali

Reputation: 2454

Late but i hope this could help you.

Check the last sections of scripts docs

  1. Be sure that your zabbix server has connection with your zabbix agent (normally cause by firewalls)
    1.1 by port 10050 for passive checks
    1.2 by port 10051 for active checks
# you can do it with telnet from your zabbix server
> telnet <agent ip> <10050 or 10051>
    Trying <agent ip>...
    Connected to <agent ip>.
    Escape character is '^]'.
  1. You can modify your server/agent config file to increase Timeout directive. By default is 3 and you can set it up to 30 seconds. If you do this, be sure to modify in both server and agent.
    2.1 Don't forget restarting the services service zabbix-agent restart and service zabbix-server restart

Upvotes: 2

Steven Sedory
Steven Sedory

Reputation: 11

after many an hour, this is what fixed it.

Active Agent Checks were failing with following error or similar: active check configuration update from [zabbix.verticalcomputers.com:10051] started to fail (ZBX_TCP_READ() timed out) For whatever reason, active agents wont be able to connect properly (active checks won't work, only passive), if server is behind a firewall NAT'd, and you don't have the following in your ifcfg-eth0 (or whatever NIC) file. It will work if you bypass the firewall and put a public IP right on the zabbix server. NM_CONTROLLED=no BOOTPROTO=static

If you use the CentOS 7 wizard, or nmtui to config your NIC, instead of manually, those lines don't get added. I noticed this because when running "ip add", I'd get the following:

2: eth0: mtu 1500 qdisc mq state UP group default qlen 1000 link/ether 00:15:5d:07:82:07 brd ff:ff:ff:ff:ff:ff inet 10.32.2.25/24 brd 10.32.2.255 scope global noprefixroute eth0 valid_lft forever preferred_lft forever

Notice the "noprefixroute". That was unsettling, so I dug for a long time online, with no leads. After adding the two lines to the NIC config mentioned above, and restarting the network, now looks like this: 2: eth0: mtu 1500 qdisc mq state UP group default qlen 1000 link/ether 00:15:5d:07:82:07 brd ff:ff:ff:ff:ff:ff inet 10.32.2.25/24 brd 10.32.2.255 scope global eth0 valid_lft forever preferred_lft forever

Upvotes: 1

Richlv
Richlv

Reputation: 4153

This errors out on the network level.

  • check that the local firewall on the Zabbix server allows incoming connections on the server port (10051 by default)
  • check that the VM network connectivity is correct

As a simple test, you can telnet from the box with zabbix_sender to the Zabbix server on port 10051. If that fails, you have a basic network issue.

Upvotes: 2

Related Questions