LightningCrash
2[H]4U
- Joined
- Dec 29, 2000
- Messages
- 2,470
Hi guys. I had a problem this weekend and I want to try to get some troubleshooting ideas before I give it another go.
Host: ESXi 5.5u1
Mobo: Supermicro X7SBE
NIC: Intel® 82573V + Intel® 82573L (Yes I had to roll that in to 5.5)
Old Modem: Moto SB5121
New Modem: Moto SB6121
Guest: Sophos UTM 9
Guest try#2: CentOS 7
I have a vSwitch configured for the RED interface, using the second NIC. It is dedicated to the cable modem and the UTM VM.
On the SB5121, everything works just fine.
When I switch to the SB6121, and the modem is provisioned, the UTM VM doesn't get a DHCP offer.
Before the modem was provisioned, I got a 172.x address and the ISP captive portal "Please call us to configure your modem."
I can plug the SB6121 into my laptop and it works just fine. Plug back into the ESXi host, no joy.
I ran tcpdump on the VM and I see a dhcp request go out on the vnic, but it never receives a response. I swapped in a CentOS 7 VM and got the same issue.
I restart the modem after almost every VM or host change.
What I have tried on the ESXi side:
I'd like to get some recommendations for things to try on the next go around, which might be this weekend or so. Any ideas?
My to-do list so far:
Thank you in advance for your ideas.
Host: ESXi 5.5u1
Mobo: Supermicro X7SBE
NIC: Intel® 82573V + Intel® 82573L (Yes I had to roll that in to 5.5)
Old Modem: Moto SB5121
New Modem: Moto SB6121
Guest: Sophos UTM 9
Guest try#2: CentOS 7
I have a vSwitch configured for the RED interface, using the second NIC. It is dedicated to the cable modem and the UTM VM.
On the SB5121, everything works just fine.
When I switch to the SB6121, and the modem is provisioned, the UTM VM doesn't get a DHCP offer.
Before the modem was provisioned, I got a 172.x address and the ISP captive portal "Please call us to configure your modem."
I can plug the SB6121 into my laptop and it works just fine. Plug back into the ESXi host, no joy.
I ran tcpdump on the VM and I see a dhcp request go out on the vnic, but it never receives a response. I swapped in a CentOS 7 VM and got the same issue.
I restart the modem after almost every VM or host change.
What I have tried on the ESXi side:
- Restarting the ESXi host
- Restarting the VMs
- Tried out a CentOS 7 VM on that interface
- Monkeyed with promiscuous mode on the vswitch and port group.
- Set the MAC of the VM VNIC to be the MAC of the physical NIC
- Moved the VM's second VNIC temporarily to the inside LAN, which got an IP from DHCP.
I'd like to get some recommendations for things to try on the next go around, which might be this weekend or so. Any ideas?
My to-do list so far:
- Bug the cable company more about it (I got the captive portal page, so I got DHCP once, right?)
- Throw a guad GbE NIC in a machine, make a bridge from 2 ports and run tcpdump on the interfaces.
- Use a quad GbE NIC in the ESXi machine instead of the unsupported onboard NIC (doubt this will help but it's trivial to try)
Thank you in advance for your ideas.