About 6 months ago we setup a new Dell R720 and configured it using Server 2012R2 Hyper-V hosting 4 VM’s, each is Server 2012R2 Std. There are 2 physical NIC’s, Broadcom 5720 QP and a Broadcom 5720 DP. The Quad NIC is teamed, the Dual NIC is teamed and each is a virtual switch. The Development VM is attached to the Dual virtual switch, which is on a subnet separate different from the Quad NIC team/virtual switch.
About a month ago we moved some development websites to the Development VM along with our intranet. Since then we are having to reboot the host machine almost daily because the Development VM with intranet and development websites is not allowing any traffic out through the NIC/virtual switch. We notice this by the intranet is unavailable. When this happens, we are unable to ping out to the other VM’s, but we can ping other computers in our domain, be we can the Development from the other VM’s on the Quad Virtual switch and can ping from other computers in the domain. The research we have done seems to indicate the VM Queues property for the Broadcom NIC ports needs to be set to “disable” for each port. However, it does not seem to make a difference if it is set to “disable” or “enable”, the result is the same, the intranet freezes after a day or so, then a reboot of the Host machine is required. Has anyone else had this same dilemma?