Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Question
Saturday, April 2, 2016 11:41 PM
I have set up my Servers following this blog:
Workgroup-and-multi-domain-clusters-in-windows-server-2016/
Node1 is HYPVCORE (Hyper-V Server 2016 TP4)
Node2 is HYPVGUI (Server 2016 TP 4 with Hyper-V Role)
On GUI-Server I created Primary DNS Suffix as per blog (failover.priv); on the non-GUI server, I used:
Set-ItemProperty "HKLM:\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\" -Name Domain -Value "failover.priv"
Ifconfig /all shows corresponding primary DNS suffix (on both servers).
Workgroup is WORKGROUP
I try to create the cluster with this:
new-cluster -name Failover -node HYPVGUI,HYPVCORE -AdministrativeAccessPoint DNS
But I always get this error:
new-cluster : There was an error adding node 'HYPVGUI' to the cluster
The node cannot be contacted. Ensure that the node is powered on and is connected to the network.
At line:1 char:1
+ new-cluster -name HowToFailO -node HYPVCORE,HYPVGUI -AdministrativeAc ...
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ CategoryInfo : ConnectionError: (:) [New-Cluster], ClusterCmdletException
+ FullyQualifiedErrorId : ClusterNodeNotReachable,Microsoft.FailoverClusters.PowerShell.NewClusterCommand
However, I can ping the servers just fine!
Can even add HYPVCORE in Server Manager in HYPVGUI.
Tried both built-in "Administrator" account (same password on both machines) as well as manually added local administrator accounts. Did also set the LocalAccountTokenFilterPolicy 1.
Did also disable the firewall on HYPVGUI and tried to create cluster from HYPVCORE to no avail.
If I run test-cluster only on the local node (test-cluster -node LOCAL), the test passes except for AD-warnings:
PS C:\Users\Administrator> test-cluster -node hypvcore
WARNING: System Configuration - Validate Active Directory Configuration: The test reported some warnings..
WARNING:
Test Result:
ClusterConditionallyApproved
Testing has completed successfully. The configuration appears to be suitable for clustering. However, you should
review the report because it may contain warnings which you should address to attain the highest availability.
Test report file path: C:\Users\Administrator\AppData\Local\Temp\1\Validation Report 2016.04.03 At 06.54.41.xml.htm
Mode LastWriteTime Length Name
-a 03.04.2016 06:54 347234 Validation Report 2016.04.03 At 06.54.41.xml.htm
If I set up a single-node cluster on HYPVCORE, I can even connect to it over its static IP address from HYPVGUI Failover-Manager.
Any ideas?
All replies (4)
Sunday, April 3, 2016 7:43 PM âś…Answered | 1 vote
Yeah, just ignore me so that I have to figure it out by myself... ;-)
What our guide is not mentioning, but is a REQUIREMENT for it to work:
You need to add the nodes with their primary DNS-Suffix, i.e. in my case
new-cluster -name Failover -node "HYPVCORE.failover.priv,HYPVGUI.failover.priv" -AdministrativeAccessPoint DNS
Also, I had to edit my hosts files to allocate the relevant FQDN hostnames to the relevant (static) IPs.
While I was at it, I also set the FQDN as trustedhosts in the server, where I invoke the new-cluster command:
set-item wsman:\localhost\client\trustedhosts -value "HYPVCORE.failover.priv,HYPVGUI.failover.priv" -Force
EDIT: originally I also set the following firewall rules on my Nodes, but further testing revealed that this is NOT required.
Set-NetFirewallRule -DisplayName "File and Printer Sharing (Echo Request - ICMPv4-In)" -Profile Any -Enabled True -Direction Inbound -Action Allow
Set-NetFirewallRule -DisplayName "File and Printer Sharing (Echo Request - ICMPv6-In)" -Profile Any -Enabled True -Direction Inbound -Action Allow
Set-NetFirewallRule -DisplayName "Failover Cluster (UDP-In)" -Profile Any -Enabled True -Direction Inbound -Action Allow
Set-NetFirewallRule -DisplayName "Failover Cluster (TCP-In)" -Profile Any -Enabled True -Direction Inbound -Action Allow
(However, in my quest I stumbled over THIS GUY who seems to be a genius (at least to the likes of me) and helps you to set up a cluster with nano servers...)
Sunday, April 3, 2016 1:38 PM
Update: I had succeeded some time ago creating a "non-domain joined" cluster and one of the nodes is still alive and kickin'.
The problem I described ago popped up when I tried to set up a new cluster with 2 fresh installs.
So for further testing purposes I tried to create a cluster consisting of one fresh install and my "old node". When I did created the cluster using the powershell on the FRESH NODE (i.e. adding the old node "remotely") it worked. When I tried it the other way around (i.e. invoking the command on the old node, adding the fresh install "remotely") it did NOT work.
Bite me, but I have no clue anymore which settings I used / changed at the old node. I do remember though, that I set up hyper-v workgroup replication before creating the cluster with this Guide/FAQ. But what of that may be required to enable the Failover Cluster? I am lost.
Monday, April 4, 2016 4:43 AM
Hi Besterino,
Glad to see you have fixed the issue by yourself, and it's kind of you to post the detailed information here. It may be of help to others have a similar scenario with you. Thanks for your sharing.
Best Regards,
Anne
Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Support, contact [email protected].
Tuesday, January 10, 2017 9:04 PM
Thanks; I was trying to get this running between two HYPER-V CORE 2016 servers (two VMs inside Hyper-V on a Windows 10 desktop, actually, as a proof-of-concept).
The last piece I required was making sure I'd added a HOSTS file entry for the other node onto the server from which I was running the `new-cluster` command.
Process for doing that was at the command prompt:
Set file="%windir%\System32\drivers\etc\hosts"
echo 192.168.1.147 HYPERV1 >> %file%
echo 192.168.1.148 HYPERV2 >> %file%