Hello all,
I am having a widespread issue where Linux VMs display tx hang error in the console and fall off the network. The issue is present regardless of the OS ( Red Hat, CentOS, Ubuntu) or the VMware Tools version installed. Most of them had old "Flexible" or E1000 nics, so we upgraded them and installed the VMXNET3. Since we did that, we still see the error in the console but the VMs still ping. While they still ping, internal processes do experience network outages. For example, we have a DNS server that stops responding to requests and logs the errors. The netdev watchdog resets eth0 and functionality comes back.
We can also restore functionality by migrating the VM to another host.
Our hosts run ESX 6.0.0 U01a 3073146 . So far VMware support does not find a cause. I wiped one of the servers and installed the new ESX 6.0.0 U01b 3380124. After initial elation and partying we see the error persist.
Our hosts are Dell R720XD and Dell R730XD with QLogic Nics.
So far VMware support wasn't able to figure it out.
Does anyone know anything at all about this?
Thank you!
