KVM guest to guest 'No data received from host'

[scottthepotter]

I was recently running Nagios 4.08 on a CentOS 6.6 VM under VirtualBox monitoring Windows servers using NSClient ++ and all was fine. I have moved the VM to a KVM host, maintaining the same MAC Address and IP address and now I am receiving 'No data received from host' from all other guest machines on the same host as the nagios guest. I am able to ping both ways from guest to guest. SELINUX is disabled. Since network traffic on QEMU bridged machines does not get intercepted by the host, I can't imagine that it would be a firewall rule. Where to start?

[sreinhardt]

Let's start with an nmap and make sure those ports are available.

Code: Select all

nmap -p 5666,12489 [remote host\ip]

[scottthepotter]

Sorry, didn't get the email that there was a reply.

Starting Nmap 6.40 ( http://nmap.org ) at 2015-01-15 14:55 MST
setup_target: failed to determine route to 12489 (0.0.48.201)
Nmap scan report for 192.168.254.117
Host is up (0.00022s latency).
PORT STATE SERVICE
5666/tcp open nrpe
MAC Address: 52:54:00:C7:5F:DB (QEMU Virtual NIC)

Nmap done: 1 IP address (1 host up) scanned in 0.09 seconds

[tmcdonald]

Looks like there was a typo in your command. Make sure there is a comma and not a space between 5666 and 12489:

Code: Select all

nmap -p 5666,12489 192.168.254.117

[scottthepotter]

Not only did I put in a space, but I realized I tried this from the Host not the guest that is running Nagios

Starting Nmap 6.47 ( http://nmap.org ) at 2015-01-16 09:12 MST
Nmap scan report for shelter-03.hslc (192.168.0.21)
Host is up (0.00058s latency).
PORT STATE SERVICE
5666/tcp open nrpe
12489/tcp open unknown
MAC Address: 52:54:00:05:09:4B (QEMU Virtual NIC)

Nmap done: 1 IP address (1 host up) scanned in 0.04 seconds

[tmcdonald]

Actually you do need to run this from the Nagios machine, targeting the remote machine. Is that what you did in your most recent post?

[scottthepotter]

Yes.

[lmiltchev]

Can you also run the following command on the nagios server from within the plugins directory and show us the output? Hide sensitive info.

Code: Select all

./check_nt -H <client ip> -s <password> -p 12489 -v CLIENTVERSION
./check_nrpe -H <client ip>

BTW, have you checked the nsclient.log for clues?

[scottthepotter]

Not running NRPE, but get the same message when I check the 'check_nt':

No data was received from host!
could not fetch information from server

Strange thing is that nagios reports that the 'host is up' so it is able to recognize the existence of the windows server, just can't pass data. Really strange as all I did was to move the nagios guest from one host to another. Although the nagios guest is on the same host as the target guests.

I double checked the configuration file on the target windows server to be sure the password is correct and it is (should be as it hasn't moved
I went ahead and allowed both ports in iptables on the host (not that it should make a difference), of course that didn't help.

[scottthepotter]

Here is the result from the nsclient log file:

2015-01-16 12:52:37: e:D:\source\nscp\trunk\include\socket/connection.hpp:134: Failed to read data: The I/O operation has been aborted because of either a thread exit or an application request

rather cryptic (at least for me).