Hello All,
Curious if anyone has come across this issue and knows of a fix. Currently we are using a pool with 3 hosts. The physical servers in this pool have four 10 gig NIC Interfaces and no RJ45 ports. two of the four interfaces on each server are bonded together with the active-active configuration and plugged into switches. These bonded interfaces are used for all VM and storage/backup traffic. Due to our limitations with out network switches we have, there are no more 10 gig ports available, So the pool management is connected via an adapter that plugs into the 10 gig fiber ports on each server and converts to an RJ45.
The issues that occurs is something happens to that adapter, which causes the port to go down, i think the device might be over heating, still not sure what is the root cause, but when this happens the host with the issue will no longer report in because the management communication is now down, VM's still stay online, but cant be managed. If I attempt to replace the adapter the issue still occurs. Only fix I have found is rebooted the host, But I have to manually shutdown all of the VM's on the host in question, then preform a host reboot.
I know that our setup isn't ideal, once we get a fiber switch, or more 10 gig ports to work with, we can put the management communication directly on the 10gig, and no longer have this issue. But curious if there is a command or something that can be done to enable or reset the interface when this happens.
Has anyone come across something like this and found a solution other then rebooting the host?
Thanks