Strange IP connection attempted by the GUI

[sigxcpu]

Hi,

I've received a strange GUI error when clicking on a VM.

The error screenshot is attached.

On the node (enc11-c03), a tcpdump shows these:

14:53:55.306867 IP 10.10.11.103.46688 > 216.21.239.197.8006: Flags , seq 3411620057, win 14600, options [mss 1460,sackOK,TS val 6765664 ecr 0,nop,wscale 7], length 0
14:53:58.306949 IP 10.10.11.103.46688 > 216.21.239.197.8006: Flags , seq 3411620057, win 14600, options [mss 1460,sackOK,TS val 6768664 ecr 0,nop,wscale 7], length 0
14:54:04.306798 IP 10.10.11.103.46688 > 216.21.239.197.8006: Flags , seq 3411620057, win 14600, options [mss 1460,sackOK,TS val 6774664 ecr 0,nop,wscale 7], length 0

A traceroute to that IP shows it goes nowhere (looks like it is not even allocated).

Does anybody have an idea on what is going on?

 

[e100]

The VM seems to be assigned to the node named enc11-c02
In your screenshot that node is not listed, only c03-c07

I assume your are logged into the GUI on node enc11-c03
When you want to manage a VM that is on another node, the node your are logged into needs to talk to the node managing the VM.
In this case that is node enc11-c02 which enc11-c03 thinks is at the IP address 216.21.239.197

What happened to enc11-c02?

 

[sigxcpu]

Hi,

Thanks for looking it up. I've found the issue.

Sometimes the UI back-end prefers to use DNS to access the hosts instead of IP addresses (I don't know why).
The DNS was not configured properly (as everything worked almost all the time and I was too lazy) and the upstream DNS belongs to some spam registrar (aren't they all?) that does catchall.

When I was trying to resolve enc11-cXX I've got that IP address and that is how I've managed to understand what was going on.
Now I have a properly configured DNS and this should not happen again.

I think we can delete this thread because the information (the catchall IP address) is meaningless, so it spams Google with fake issues

Thanks again,
Alex