Cluster NIC failure

[Simcfc73]

Evening,

 

I was doing some upgrades on my network (inherited.. still finding weird stuff) and to cut a long story short I've had to roll back to where I was but it messed up my cluster. 

I managed to get 1 node up but the other 1 has stopped. Somehow the NIC's all went funny so the array NIC won't enable. I'm sure I'm missing something obvious.

 

I don't have the original config of how its setup annoyingly.

 

I managed to get the DC up which isn't on the array. It's on Hyper-V but that virtual switch was in error also so had to remake one. 

 

I managed to get a NIC working with a IP but can't connect to the server. (I've had to remote into the iDrac to sort it)

 

That make any sense?

[synaesthesia]

If multiple had gone it's likely a teamed NIC issue which isn't uncommon with driver updates. Also what's the host OS? 2022/2025 also have deprecated NIC teaming via the server manager, so wonder if an update might have enforced that?  You now need to manage those teams with powershell (SET teams? Only did ours a couple weeks ago and already forgotten!)

That will certainly include clustered configurations. Would advise removing all teams and resetting them up as switch embedded teams - quick reference at https://learn.microsoft.com/en-us/azure/azure-local/concepts/host-network-requirements?view=azloc-2601#switch-embedded-teaming-set?context=/windows-server/context/windows-server-virtualization

[DalekSec]

I had an issue back when 2022 was released with teaming, make sure you're doing the changes via Powershell and not the GUI, the GUI made ours blue screen when set!

[Simcfc73]

Thanks for this.

 

Just got in and my desktop rebooted and the toolbar won't come back on. Happy fridays

[Simcfc73]

I'm having to learn all this as I've never used clusters before.

 

It's a 2019 server and looking at the back there's 2 fibres in each with one of them linked together and the other 1s into the switch.

 

I can't ping either of them but I've got 1 of the clusters on.. the other says no and not showing anything. They can't ping each other either.

The second one that's got a failed cluster has a disabled nic which won't enable.

Two of the hyperv clients are running but won't connect to the network. They use the same virtual switch.

 

Not sure what the best way forward is without downtime. I really wanted to update to 2025 tbh while I've got this like it is. I was thinking of migrating everything to one nose and upgrading but they won't talk to each other.