I need your help in understanding why I'm facing this strange problem.

So I'll get to the set up:

Windows 10 laptop as a host let's call this host win10. Vmware installed on win10

two Linux VMs on VMware:

  1. fedora desktop let's call this vm1

  2. fedora server let's call this vm2

both vm's are configured with static IP, somehow they both are automatically using the VMware application as a dns server which, as mentioned somewhere in vmware documentation, uses the win10 to resolve names.

win10 has it's hosts files appended with the ip mappings of both vm's

Now here comes the twist:

Another device is my android phone which is running wifi for win10 to connect to. Both vm's use NAT to access the internet through my phone.


When win10 is connected to android wifi. All works well!!!

When disconnected however both vm's can nslookup each other and vm1 can ping both itself, vm2 and win10. Vm2, however, can't ping anything it seems not even itself.

Please provide any advice to solve this issue. I have no idea why everything works fine when connected to the internet but not when disconnected. None of my devices should depend on the internet for name resolution of local resources!!!

I did some research and I found out that maybe I have this avahi software on vm1. But I'm not sure whether this can be the difference. Still doesn't explain why I can't ping with my vm2.


Check your routing (route -n) on each vm, when connected to the internet and when not. They may both be able to reach a local nameserver but if vm2 has a lower-weight default route on a non-routeable interface, all non-local traffic including pings will get lost.

It would help to see what IP addresses it has on each interface. And whether there are any address collisions.

(Some of this is for clarification but at the time I did not have enough reputation here to add a comment.)

| improve this answer | |
  • I would like to clarify my question more. In all situations, I can ping all hosts from each other using IP. It's when I use hostnames that I have trouble. – admdev Jun 2 '17 at 12:30
  • OK, I've edited my answer to remove the point about local pings. It would still be useful to see the IP address and route -n output. – Joe P Jun 2 '17 at 13:33
  • I have been trying to reproduce the problem but no success. everything seems to be working now. I will post again once I see this happening again. Thank you. – admdev Jun 2 '17 at 14:52
  • The problem is back and I have to correct my observation that vm2 can't ping itself. Actually it can. but nothing else. here's route -n output from vm1: line 1: ug 100 0 0 enoxxx line 2: u 100 0 0 enoxxxxx. Vm2 similar to vm2 except different Iface. – admdev Jun 3 '17 at 12:57
  • It would be good to edit your question and put in these extra details. In fact much better, because then you can format it. – Joe P Jun 4 '17 at 20:08

I posted this question a while ago and I think it's worth it to mention that I re-installed VMware and now things are behaving differently. It's also worth to mention that there are two configuration files that one could look into in a situation like this:

  1. C:\ProgramData\VMware\vmnetdhcp.conf

  2. C:\ProgramData\VMware\vmnetnat.conf

After changing configuration in these files, you need to restart the corresponding services in Windows for the change to take effect.

This isn't necessarily an answer to my question but I thought this could help someone in a similar situation.

| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.