Matt Jones has provided another excellent article with problems and solutions at Gorillapond.com
This time he talks about DHCP server problems inside virtual machines, he says in the article "It appears to be an issue with DHCP requests from different VLANS to a virtual machine running a DHCP server." He goes on to explain the solution which is just to reset the virtual network card.
For the last 2 days we were having network connectivity trouble with our clients. It appears to be an issue with DHCP requests from different VLANS to a virtual machine running a DHCP server. I’m not sure it’s caused by ESX’s virtual networking or ESX’s interaction with our Cisco equipment. I have had this issue with both Windows 2000 Server and Windows Server 2003’s DHCP servers. It would be interesting to know if a Linux server would have the same issue. I have had this issue a few times, once when we first setup our virtual infrastructure and again just this week after I rebooted the DHCP server.
Problem:
Clients that were rebooted were not able to logon properly or surf the internet, they effectively dropped off the network. Clients were receiving IP addresses that were not for their VLAN. For example, clients in the 10.80.8.x/21 subnet were getting 10.80.0.x/21 addresses. These clients were only able to ping the DHCP server.
Solution:
- Open Virtual Infrastructure Client
- Find & select virtual machine running DHCP
- “Edit Settings” on virtual machine
- Select network adapter used for DHCP
- Uncheck “Connected”
- Click “OK”
- “Edit Settings” on virtual machine again
- Select network adapter used for DHCP
- Check “Connected”
- Click “OK”
Essentially you are resetting the network connection. I don’t know why this works, but I’d like to know!
