NIC error: "not a valid pnic 'vmpnic0' for teaming on portgroup"

[virtualesxer]

After having added a PCI card. my ESX 3.5 server reboots to incorporate the hardware changes. So far so good. Then, when it has started again, I seem to have lost my NIC somehow. Ok, just remove the new PCI card, right? I did, figuring might be some IRQ conflict. But still, "host unreachable" in all directions.

Now, mind you, I'm very new to ESX, so I had hardly even a clue where to start troubleshouting this. But googling I found a few useful esxcfg commands, and learnt the following:

1) esxcfg-nics -l lists the Intel NIC (E1000) as 'up' and running at full speed, full duplex. So, no hardware malfuction.

2) Grepping /var/log/messages, I see errors like "not a valid pnic 'vmpnic0' for teaming on portgroup". Also, I noticed my NIC is apparently listed/configured (output of esxcfg-nics -l) as 'vmpic3'. So, I'm thinking, maybe vmware rearranged the NIC-name somehow, but not all the way (after I added another PCI card). If so, then the question is, of course, where do I go to change it back? I wouldn't begin to know where ESX stores that info. Or is there some sort of network rediscovery command to get vmware to get things right again?

Of course, the above may just have been wishful thinking on my part, but it looks like there may be an easy fix to this.

Thanks

[dmaster]

some usefull network settings files. (after chnaging this file type : service network restart)

/etc/sysconfig/network

/etc/sysconfig/network-scripts/ifcfg-vswif0

-

---

what output do you get with the following commands:

esxcfg-vswif -l

and

esxcfg-vswitch -l

after this you can quickly see, most of the times what the problem is..

in your scenario something happened / changed with the service console network.

see if you still have a vmnic assigned to the service console portgroup and that the link is up.

root@ESX1 root]# esxcfg-vswitch -l

Switch Name Num Ports Used Ports Configured Ports MTU Uplinks

vSwitch0 64 4 64 1500 vmnic0

PortGroup Name VLAN ID Used Ports Uplinks

Service Console 0 1 vmnic0

[depping · Accepted solution]

ESX indeed rearranges the vmnic names sometimes when you add a PCI Card. This can result in these kind of problems. I would suggest unloading all nics from your vswitches and re-adding them in the right fashion.I usually unplug all nics and plug them in one by one and just write down which vmnic is which physical nic.

Duncan

My virtualisation blog:

If you find this information useful, please award points for "correct" or "helpful".

[virtualesxer]

ESX indeed rearranges the vmnic names sometimes when you add a PCI Card. This can result in these kind of problems. I would suggest unloading all nics from your vswitches and re-adding them in the right fashion.I usually unplug all nics and plug them in one by one and just write down which vmnic is which physical nic.[/i]

That's it! SOLVED! Apparently, the uplink was removed from the switch. Using the 'vmnic3' label I found earlier, I did this:

esxcfg-vswitch vSwitch0 -Lvmnic3

And I was back in business. Pfew, it would have taken me ages to find that out! So, thanks again!

[virtualesxer]

ESX indeed rearranges the vmnic names sometimes when you add a PCI Card. This can result in these kind of problems. I would suggest unloading all nics from your vswitches and re-adding them in the right fashion.I usually unplug all nics and plug them in one by one and just write down which vmnic is which physical nic.

I got one question left on this. I'm planning to add a second Intel PRO/1000 GT Desktop Adapter PWLA8391GT (like the one I have now). Will I be able to assign that card to a single VM, exclusively? And is adding it likely to completely confuse ESX again? (I don't understand why it should become so confused in the first place).

Thank you.

[JoeByrne]

Thank you very much. Had upgraded my bios and firmware yesterday on a Dell R900 and added a Sas card then bang, rebooted no network. Pretty new to VM configuration so was very happy to find this solution. Saved me hours!