• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Strange Vmware connection issue

P

piker28

Limp Gawd
Joined
Aug 2, 2007
Messages
183
Hello all, so on one of my hosts I am having strange connection issues with my vm's that are running. I currently have 7 NIC's hooked up and 8 vm's running.

What I noticed the other day is 3 out of the 8 vm's will not connect to the network. If I reboot the host and the vm's; the vm's that connect will change. So for example server's 1,2,3 will connect now but if I restart the host those will no longer connect.

All the servers are running on static ip's and do not have a conflict on the network.

I hope that was not to confusing, but does anyone have any ideas on how to solve this riddle?
 
how are the nics hooked up? What version of VMware software? What kind of NICs? Do you have logs from the host?
 
To add to Lopoetve's request:

What ESXi version?

Standard switch or distributed switch?
Port group load balance policy?
vSwitch vmnic settings (active vs standby)
-- if standard switch, what is the "Number of Ports" configured to?

Any VM mac address conflicts?
Do the VMs pass through an external router to communicate (possible mac address table limit?)
Linked Clones?


Nick
 
Sorry for the late reply. In case if anyone is wondering I believe I have solved the mystery. The servers having the issues are actually running in a cluster setup. Well the idiot who set it up never actually licensed it so once the trial expired they apparently locked them off.
 
Sorry to bring this back up but it appears I was wrong all along.

This issue has crept back up to our host and even shows signs in out second host as well.

The ESXi version is 5.1.0

The NIC's are hooked up in a standard switch. It is setup for 120 ports. All adapters are in the active.

So for the VM network it uses 1,2,5 for the active adapters, 6 for the standby, and 0,7 for the unused.

For the Management network it uses 0,7 for the active, 6 for standby and 1,2,5 for the unused.

I do not see any mac address conflicts. They connect directly to the cisco switch and do not pass any other device.


When I remove all the teaming and hook up one NIC, everything starts to work fine.
 
I have seen this happen, when the switch is not handling the data correctly from several ports bonded like this. I would see routing errors on the switch and a failed port assignment, that might be what is going on here.
 
As in the vm switch or the physical switch?

I did more testing and when I have one adapter in the active everything is happy, whenever I add another adapter then things start to stop working again.
 
Under the host--> Under Configuration --> Under Networking

Open vSwitch0

You need to examine the Failover and Load Balancing settings

Also check the Network Failure Detection states and Failback state settings.
 
You made ports 1,2,5 sound like they should always be active, if the physical switch does not know what to do with the data on multiple ports things go haywire
 
Yea this was the setup of the person who put this in place. I admit I am new to the larger VM setups. So it was working for months before hand.
 
You must log in or register to reply here.
Back
Top