Hi,
I'm running v6.7 and have found something odd and not sure what the next steps would be.
Scenario:
1. Create a Server 2016 or 2019 VM from template (on any of my 3 hosts)
2. Join to domain (and reboot)
3. Apply a static IP
4. The network connection will work for 10-30 minutes, and then stop working.
- Unable to ping out (via IP), and it can't hit the gateway
- IP info is still there
- If you disable the NIC and then enable the NIC, it works ok for another 10-30 minutes
- If I vMotion it to another host, disable the NIC, enable the NIC, it then seems to work fine and I can not recreate the issue (on that VM).
Is this a bug? Anything to look at to try to resolve?
Do you have the VMware tools installed? What type of NIC are you using: E1000 or VMXNET3?
The behavior sounds like a duplicate IP problem to me. Did you check that?
Hi,
Welcome to VMTN. ![]()
Have you tried TCP/IP stack reset ?
Check this blog link: Truly reset the TCP/IP stack
Edit: Formatting
VMWare Tools is installed and it's using the VMXNet3
It honestly sounded like that to me too.
I can reproduce it and it's happened to over 10 VM's.
If the machine is NOT on the domain but keeps the same static IP, it's fine. It's once I join it to the domain, it'll randomly drop. If I turn DHCP on once on the domain, it'll also be fine.
To be clear, it will work for up to ~30 minutes with a static IP and on the domain.
What's the best way to do this?
I think my biggest problem is that I don't want to do this for every new machine I spin up?
Try this once on a VM (deployed from template) and monitor it for few days. If this resolves the issue, convert the template into VM, perform same steps on it and convert the VM back into template.
Then deploy a new VM from updated template and it should work fine.
Unfortunately, this didn't work.
I did find out that if I deploy the machine from the template, delete the NIC, add the same vNIC back (same config/vlan etc), I don't have the problem on the domain + static IP.
You can test your finding on the template and the new VMs deployed from same template should not face the issue anymore. Let me know how it goes.
