How to access vms in a failover cluster?

[danswartz]

I'm not sure how to proceed here. 2 ws2019 hosts in hyper-v role. LAN is 10.0.0.0/24 and storage (to JBOD) is 192.168.3.0/24 (10gb enet). I noticed slow backup and realized I had added the 2 hosts by their LAN names. I removed them, and added back by their 10gb names, and got much better speed. So far, so good. The puzzle: I have the cluster up and running with the 2 nodes added, but no VM roles yet (planning on using starwind VSAN, but haven't gotten to that yet). The issue: once I have the vms moved to the cluster, I will have to add the vms from the cluster, and then remove the 2 standalone hosts from veeam b&r. The problem: the cluster and active directory all know the 2 nodes by their LAN ip addresses, so it seems to me that if I don't change something, I will be back to 1gb backup speeds, no? I guess I can change the cluster IP to the 10gb subnet, evict the 2 nodes and add them back by their 10gb names, but I'm wondering if there is an easier way? Thanks!

p.s. this is B&R 12.1 just installed a few days ago - community edition.

[Mildur]

Hello Dan

I'm not a Hyper-V expert. But wouldn't it better to have storage and production LAN isolated from each other?
Maybe you can update the network adapter for your LAN to 10 Gbit as well.

Best,
Fabian

[danswartz]

This all turned out to be some kind of misconfiguration between veeam and hyper-v. All set now, sorry for wasted bandwidth.

[supa_plex]

Hi Dan,
Nice to hear that you've found the misconfig. JFYI, you can configure preferred networks in Veeam, which will be used for Veeam traffic between hypervisor hosts and backup repo.
https://helpcenter.veeam.com/docs/backu ... ml?ver=120
From my experience VM or server with Veeam should be rebooted after you apply the rules (rules didn't work without that in my case). The following thread might also help:
veeam-backup-replication-f2/backups-not ... 73118.html

[danswartz]

This was truly mysterious. I got everything working, then, after a few hours, backup and replication jobs would fail with a message about Hyper-V integration not being accessible from/on the host. Rescan of hosts&cluster would 'fix' this but only briefly. I was on the veeam physical host and happened to ping one of the hypervisors, and was surprised to see it was responding but not on the IPv4 address! e.g. I was doing something like 'ping hyper1'. I then examined the network settings and confirmed that both hosts had the correct IPv4 static IP addresses. On a hunch, I looked at the DNS page off the network settings, and saw there was no default domain listed, so I added my domain, and rebooted both hosts, and confirmed that 'ping hyper1' and 'ping hyper2' behaved as expected. This was several days ago, and I have yet to have another failure. The OCD in me would love to know what the **** was going on, but it works now, so...