How to unregister Prism Element/Cluster from Prism Central
There are various scenarios in which you may need to unregister a cluster from PC, and it is important to do it correctly.
Whether you are decommissioning a Prism Element (PE) cluster which was registered to Prism Central (PC), or you already have decommissioned a cluster but it is still linked to a PC, or you have a cluster that is registered with one PC instance but would like to re-register it with a different PC instance for the benefit of localized management or to configure availability groups using Leap. All
What does “correctly” look like?
Done properly, unregistration of a cluster from PC involves a remove-from-multicluster step followed by clean-up of associated metadata. This metadata clean-up must be allowed to complete prior to attempting to re-register the cluster to a PC, otherwise, registration could be blocked.
How to do it?
There is no GUI method to unregister a cluster from Prism Central, so the process requires SSH access to the PC VM as well as to a CVM of the cluster. Please use “nutanix” user to authenticate with both the PC and the CVM.
Depending on the scenario at hand the process differs slightly.
1. If the PE cluster still exists and can communicate with Prism Central, you’ll need to get the UUID of the cluster from the CVM and then use that to trigger de-registration from PC command line.
2. If both the PE and PC cluster still exist but cannot communicate because the PE cluster was relocated or changed to a different network, you will need to complete the un-register and clean-up steps on both PE and PC.
If the PE cluster or the PC VM(s) was destroyed without completing these steps, you should complete unregister and clean-up steps on whichever the system remains (the cluster or the PC), following the same method as in scenario 2.
Below Solution work in all the scenario
- Take SSH session of the virtual IP or any controller VM of the registered cluster(Prism Element).
- Run the “cluster status” command and verify that all services on all nodes are healthy.
nutanix@CVM:~$ cluster statusCVM: 192.168.X.X Up, ZeusLeader
Zeus UP [14940, 14988, 14989, 14990, 14999, 15017]
Scavenger UP [16820, 16953, 16954, 16955]
Xmount UP [19228, 19314, 19315, 19330]
SysStatCollector UP [18319, 18467, 18468, 18469]
IkatProxy UP [20365, 20499, 20500, 20501]
IkatControlPlane UP [21863, 22011, 22012, 22013]
SSLTerminator UP [21949, 22077, 22078]
SecureFileSync UP [21975, 22193, 22194, 22195]
Medusa UP [22879, 23001, 23002, 23018, 23494]
DynamicRingChanger UP [23842, 23923, 23924, 23951]
Pithos UP [23873, 24082, 24083, 24092]
InsightsDB UP [23927, 24132, 24133, 24137]
Athena UP [24051, 24196, 24197, 24198]
Mercury UP [24153, 24356, 24357, 24366]
Mantle UP [24200, 24424, 24425, 24434]
VipMonitor UP [2853, 2854, 2855, 2856, 16769]
Stargate UP [26264, 26383, 26384, 26399, 26400]
InsightsDataTransfer UP [26804, 26956, 26957, 26959, 26960, 26961, 26962, 26963, 26964]
Ergon UP [26889, 27116, 27117, 27118]
GoErgon UP [27030, 27184, 27185, 27208]
Cerebro UP [27128, 27275, 27276, 27433]
Chronos UP [27143, 27378, 27379, 27387]
Curator UP [27214, 27422, 27423, 27473]
Prism UP [27470, 27675, 27676, 27696]
Hera UP [27626, 27792, 27793, 27796]
CIM UP [27688, 27886, 27887, 27892, 27893]
AlertManager UP [27848, 28019, 28020, 28104]
Arithmos UP [27926, 28158, 28159, 28201]
Catalog UP [28234, 28379, 28380, 28381]
Acropolis UP [28631, 28696, 28697, 28698]
Uhura UP [28645, 28824, 28825, 28826]
NutanixGuestTools UP [28707, 28892, 28893, 28902, 28930]
MinervaCVM UP [30114, 30250, 30251, 30252, 30698]
ClusterConfig UP [30127, 30357, 30358, 30359]
APLOSEngine UP [30255, 30426, 30427, 30428]
APLOS UP [30951, 31128, 31129, 31130]
PlacementSolver UP [31054, 31193, 31194, 31196, 31205]
Lazan UP [31082, 31289, 31290, 31291]
Polaris UP [31163, 31366, 31367, 31472]
Delphi UP [31228, 31394, 31395, 31396, 31414]
Security UP [31300, 31450, 31451, 31452]
Flow UP [31347, 31537, 31538, 31540, 31553]
Anduril UP [31487, 31671, 31672, 31673, 31687]
XTrim UP [31539, 31734, 31735, 31736]
ClusterHealth UP [989, 1015, 1016, 1134, 1135, 1137, 1138, 1140, 1141, 1152, 1155, 1159, 1160, 1162, 1167, 1169, 1183, 1219, 1222, 5362, 5395, 5396, 31595, 31776, 32115, 32116, 32121, 32131, 32181, 32182, 32185, 32188, 32189, 32210, 32211, 32227, 32228, 32234, 32235, 32245, 32340, 32341, 32348, 32349]
2024-03-12 13:25:42,186Z INFO MainThread cluster:3104 Success!
3. Get the Cluster ID of the registered Prism Central:
nutanix@NTNX-18SM62431247-C-CVM:192.168.212.11:~$ ncli multicluster get-cluster-state
Registered Cluster Count: 1Cluster Id : cbf18264-4a02-4476-a0af-a6d93436a4e6 <<< Clusterid
Cluster Name : Unnamed
Is Multicluster : true
Controller VM IP Addre... : [192.168.212.31]
External or Masqueradi... :
Cluster FQDN :
Controller VM NAT IP A... :
Marked for Removal : false
Remote Connection Exists : true4. Run the below command to remove the Prism Central from Prism Element.
nutanix@NTNX-18SM62431247-C-CVM:192.168.212.11:~$ ncli multicluster delete-cluster-state cluster-id=cbf18264-4a02-4476-a0af-a6d93436a4e6
Cluster state deleted successfully
Once the above command run successfully , We can check the Prism Element dashboard.
If you want to Unregister PE from PC
1.Take the SSH Session of the Prism Central and run the below command to get the cluster ID.
nutanix@NTNX-192-168-212-12-A-PCVM:~$ ncli multicluster get-cluster-state
Registered Cluster Count: 2
Cluster Id : 0006001b-025c-0c86-0000-0000000253c4
Cluster Name : Nutanix
Is Multicluster : false
Controller VM IP Addre... : [192.168.212.170, 192.168.212.171, 192.168.212.172]
External or Masqueradi... : 192.168.212.190
Cluster FQDN :
Controller VM NAT IP A... :
Marked for Removal : false
Remote Connection Exists : true
Cluster Id : 00060cd7-0777-8e4a-0000-0000000196b6
Cluster Name : NX-DR
Is Multicluster : false
Controller VM IP Addre... : [192.168.212.151, 192.168.212.152, 192.168.212.153]
External or Masqueradi... : 192.168.212.163
Cluster FQDN :
Controller VM NAT IP A... :
Marked for Removal : false
Remote Connection Exists : false2. Run the below command to remove the Prism Element from Prism Central.
nutanix@NTNX-192-168-212-12-A-PCVM:~$ ncli multicluster delete-cluster-state cluster-id=00060cd7-0777-8e4a-0000-0000000196b6
Cluster state deleted successfully
See also :- 1 LDAP URLs configured by IP address, please consider using FQDN.