Hello to all
Today I encountered a very strange behavior in one of my hosts: All physical adapters in a specific virtual switch do not return the CDP (or any other) information. While trying to resolve this, I also saw that these adapters do not even realize that I disconnect the cable. They look like the cable is still connected.
Any suggestions on this. I am trying hard not to restart the hosts, because it has multiple production systems on it
Thank a lot
Hi @Ody42
Is the upstream switch showing that those ports are also down?
Are those links used for any network traffic, if so are there any VM’s connected to the corresponding port groups which you could do connectivity tests from to see if the link is still passing traffic and it’s a bug in the status reporting?
Hey @Ody42,
A few questions:
Hi @dochemlock
The upstream switch does not show that these ports are down
There is, indeed, network traffic via the specific vSwitch, which was obviously interrupted during the incident. I had to migrate the VMs that have at least one interface connected with this vSwitch
Hey @Lalegre
The problem occurred in only one host, and only one vSwitch of it
The status of the physical NICs (from the switch side) looks OK
The physical ports of the switch look up/up
No, I don't have Beacon Probing configured
Do you have the NIC drivers with the same version on all the hosts?
Yes. Most probably looks like it's a firmware or hardware issue