jump to navigation

Which physical NIC is my VM using? June 23, 2011

Posted by General Zod in Tech, VMware.
trackback

Recently, I was having an issue in which some of my VMs would intermittently become unreachable across the network.  Looking in vSphere, there were no problems reported with the physical NICs nor with the virtual networking.  Everything was configured OK, so what was the problem?

Then, I noticed that all of the effected VMs were being hosted by my HOST3 server.  I started a continuous ping to one of the VMs, and then VMotion’ed the VM from HOST3 to HOST4.  As soon as the VM finished transferring to HOST4, the pings picked up and it was accessible once more.  So I’m inclined to believe that this is an issue with one of the physical NICs that’s reporting itself in a false-positive state.

Now all of the VMs in my cluster are running off of the same vNetwork Distributed Switch, and each Host has 2 physical network connections dedicated to this vSwitch… which are on different physical cards.  So which physical NIC in HOST3 could be at fault?  How exactly do you figure out which NIC your VM is communicating across?

Well… finding this out is easier than you’d think.

  • Open up your favorite emulator, and SSH into your Host.
  • Then, run the ESXTOP command, and then press “n” to switch it to the Networking Usage display.

esxtop-n What you’re presented with will be a list that includes every VM that is running on your Host server, and the currently associated “vmnic” number.  (In my case, vmnic1 is the culprit.)

Then, pull up your VI client and select the Host in question.  Go to the Configuration tab, and select Network Adapters.  The displayed “vmnic” list is grouped together by the physical card that each port is on.  network-adaptersAs you can see by the image, vmnic1 is the 2nd port on the integrated NC375i network adapter.

Guess it’s time to call HP.

Comments»

1. Ken - January 11, 2012

All of our DL580 G7s have the NC375i and NC375T. Even with the latest firmware and driver we are still seeing problems similar to what you describe. HP is aware of the problem and has been working with QLogic for a very long time to try and fix this, but even now it is not fixed. We are in the process of switching to Intel Network cards as we are tired of slowdowns and false positive issues.

2. JohnDoe - February 13, 2012

Hey Ken,

Just curious, what FW/Driver and VMWare version are you on? We have the same exact problem and am in the final stages of switching to the NC365T quads. Nice to know we are not the only one.

3. General Zod - February 14, 2012

At the time, I was using several different hardware models running ESXi 4.1 Update 1… but I believe that this issue was on a HP DL580 G7 that was flashed up to FW 9.20b. However, if memory serves me correctly… this turned out to NOT be a hardware issue.

Before I’d contacted HP, I decided to simply wipe and rebuild the Host… and I haven’t had a problem with it since. I never did find out what had caused the original problem, but it did get fixed.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: