"Zabbix agent on host.name is unreachable for 5 minutes"

[leandroceccotti]

Hi,

i've installed Zabbix Server 2.4 in Debian 7 and the agent 2.4.4 in Windows 7 hosts and no i get the following alert:

"Zabbix agent on host.name is unreachable for 5 minutes"

The log file in host give me this information:

2976:20150810:105736.991 In collect_perfstat()
2976:20150810:105736.991 End of collect_perfstat()

At Discover I see Zabbix doesn't recognize the existence of an agent on the host, but I did exactly the same in another PC and no problems ..

Any solution? Any idea?

Thanks and forgive my english!

PS: I've increased the TimeOut=30 in server.conf file and nothing.

[LenR]

Is your zabbix port open on the windows system?

On the windows system "netstat -an" to make sure your zabbix agent is listening

On the zabbix server, try telnet <windows system> 10050

or

nmap -p 10050 -P0 <windows system>

[LEMORALES]

@leandroceccotti, Hi

I have the same issue than you since yesterday. As @LenR said, I have checked ports, tested with telnet since the server and from the client and everything seems to be ok. In fact, the trigger that turns on the alert is based in an item that is returning "Up (1)", which shouldn't turn on the alert.

The item uses the key "agent.ping", type is zabbix agent, type of information is Numeric (unsigned), Data type is Decimal, the history storage period in days is 7. This is the default "agent ping" item in the default "App Zabbix Agent" template.

The trigger is as follows: {servername:agent.ping.nodata(5m)}<1. Again, this is the default trigger in the default "App Zabbix Agent" template. All the servers monitored that uses this trigger has this problem.

I am trying to fix the issue re-writing the trigger with an equivalent expression like "{servername:agent.ping.nodata(300)}#1", but zabbix keeps reporting that the zabbix agent is unreachable for 5 minutes, despite that the item "agent ping" is returning 1.

What stops the alert is re-writing the trigger as follows: {servername:agent.ping.nodata(300)}=1, which, if I am right with my logic, it has no sense regarding the fact that I want to alert when the agent.ping key returns 0.

Something very rare is the fact that I have two zabbix servers, one in 2.0.2 version and the another in 2.2.5. Each server are in separate locations, none has nothing to do which the other, and both has the same problem. Could it be a bug?. I don't know how figure out if this is a real bug to report.

Thanks and also please forgive my english.

[leandroceccotti]

Is your zabbix port open on the windows system?

On the windows system "netstat -an" to make sure your zabbix agent is listening

On the zabbix server, try telnet <windows system> 10050

or

nmap -p 10050 -P0 <windows system>

Hi, thanks for answer.

In Windows host, the port 10050 is "LISTENING"
In Server, nmap give me back this answer

root@leon:/home/adminit# nmap -p 10050 -P0 192.168.133.74

Starting Nmap 6.00 ( http://nmap.org ) at 2015-08-12 14:28 ART
Nmap done: 1 IP address (0 hosts up) scanned in 0.52 seconds

So i think all seems to be correct.

I saw in other threads that may be a false positive.

Any other idea??

Thanks for your answer.

[LenR]

Your nmap output looks too short, it should be more like this:

$ nmap -p 10050 -P0 localhost

Starting Nmap 5.51 ( http://nmap.org ) at 2015-08-12 13:42 CDT
Nmap scan report for localhost (127.0.0.1)
Host is up (0.00014s latency).
Other addresses for localhost (not scanned): 127.0.0.1
PORT STATE SERVICE
10050/tcp open unknown

Nmap done: 1 IP address (1 host up) scanned in 0.06 seconds

Which shows the nmap client can see the target has port 10050 open. (simulated on localhost for this example)

[leandroceccotti]

Resolved!!!

Hi!

I just ping from Zabbix Server FRONTEND to my Windows host and see it was pointing to another IP.

Try changing the Windows host's IP in the frontend and see what happens.

Also i changed the "RefreshActiveChecks=60" but i don´t know if this make impact.

Good luck and thanks for your help!!!!

[vikozo]

hello

i have also the Problem with

Zabbix agent on Host.name is unreachable for 5 minutes

but i do not have it all the time, after a while the error is away and come back a couple of houer later

where ist dies place to Change? on the Agent?
"RefreshActiveChecks=60"

have a nice place
vinc