Deleting NSX Controller Using API |

Today while cleaning up my lab, I came across situation where I needed to delete one of the deployed controllers. Although this task is fairly simple from vCenter UI, but recently I came across a situation (in VMware HOL) where I was unable to delete a controller via UI.

As an alternative, I came across set of API calls which did the job for me. In this post I will demonstrate how to use API calls to delete stuck/bad NSX controllers.

Step 1: Fetch controller details

# CONTENT-TYPE: application/xml
# METHOD: GET
# API CALL: https://NSX-MGR-IP/api/2.0/vdn/controller/

# CONTENT-TYPE: application/xml

# METHOD: GET

# API CALL: https://NSX-MGR-IP/api/2.0/vdn/controller/

Example: curl -sik -u “vcadmin@corp.local” -H ‘Content-Type: application/xml’ -X GET https://nsxmgr-01a.corp.local/api/2.0/vdn/controller | tidy -xml -indent -quiet

Output

<?xml version="1.0" encoding="utf-8"?>
<controllers>
  <controller>
    <objectTypeName>Controller</objectTypeName>
    <revision>0</revision>
    <name>nsx-ctlr01</name>
    <clientHandle></clientHandle>
    <isUniversal>false</isUniversal>
    <universalRevision>0</universalRevision>
    <id>controller-1</id>
    <ipAddress>192.168.109.250</ipAddress>
    <status>RUNNING</status>
    <upgradeStatus>NOT_STARTED</upgradeStatus>
    <version>6.3.7073587</version>
    <upgradeAvailable>true</upgradeAvailable>
    <virtualMachineInfo>
      <objectId>vm-48</objectId>
      <objectTypeName>VirtualMachine</objectTypeName>
      <vsmUuid>42007DBE-3706-0598-4AEE-CCD932F94B3A</vsmUuid>
      <nodeId>1b060741-08e6-42a1-9dee-b39ea845337d</nodeId>
      <revision>15</revision>
      <type>
        <typeName>VirtualMachine</typeName>
      </type>
      <name>
      NSX_Controller_78bdb508-aaea-41e3-aed9-1b10e296b323</name>
      <scope>
        <id>domain-c7</id>
        <objectTypeName>ClusterComputeResource</objectTypeName>
        <name>Compute Cluster-Site-A</name>
      </scope>
      <clientHandle></clientHandle>
      <extendedAttributes />
      <isUniversal>false</isUniversal>
      <universalRevision>0</universalRevision>
    </virtualMachineInfo>
    <hostInfo>
      <objectId>host-15</objectId>
      <objectTypeName>HostSystem</objectTypeName>
      <vsmUuid>42007DBE-3706-0598-4AEE-CCD932F94B3A</vsmUuid>
      <nodeId>1b060741-08e6-42a1-9dee-b39ea845337d</nodeId>
      <revision>61</revision>
      <type>
        <typeName>HostSystem</typeName>
      </type>
      <name>esxi-01a.corp.local</name>
      <scope>
        <id>domain-c7</id>
        <objectTypeName>ClusterComputeResource</objectTypeName>
        <name>Compute Cluster-Site-A</name>
      </scope>
      <clientHandle></clientHandle>
      <extendedAttributes />
      <isUniversal>false</isUniversal>
      <universalRevision>0</universalRevision>
    </hostInfo>
    <resourcePoolInfo>
      <objectId>resgroup-8</objectId>
      <objectTypeName>ResourcePool</objectTypeName>
      <vsmUuid>42007DBE-3706-0598-4AEE-CCD932F94B3A</vsmUuid>
      <nodeId>1b060741-08e6-42a1-9dee-b39ea845337d</nodeId>
      <revision>33</revision>
      <type>
        <typeName>ResourcePool</typeName>
      </type>
      <name>Resources</name>
      <scope>
        <id>domain-c7</id>
        <objectTypeName>ClusterComputeResource</objectTypeName>
        <name>Compute Cluster-Site-A</name>
      </scope>
      <clientHandle></clientHandle>
      <extendedAttributes />
      <isUniversal>false</isUniversal>
      <universalRevision>0</universalRevision>
    </resourcePoolInfo>
    <clusterInfo>
      <objectId>domain-c7</objectId>
      <objectTypeName>ClusterComputeResource</objectTypeName>
      <vsmUuid>42007DBE-3706-0598-4AEE-CCD932F94B3A</vsmUuid>
      <nodeId>1b060741-08e6-42a1-9dee-b39ea845337d</nodeId>
      <revision>37</revision>
      <type>
        <typeName>ClusterComputeResource</typeName>
      </type>
      <name>Compute Cluster-Site-A</name>
      <scope>
        <id>datacenter-2</id>
        <objectTypeName>Datacenter</objectTypeName>
        <name>Prod-DC01</name>
      </scope>
      <clientHandle></clientHandle>
      <extendedAttributes />
      <isUniversal>false</isUniversal>
      <universalRevision>0</universalRevision>
    </clusterInfo>
    <managedBy>42007DBE-3706-0598-4AEE-CCD932F94B3A</managedBy>
    <isDiskReadOnly>false</isDiskReadOnly>
    <datastoreInfo>
      <objectId>datastore-41</objectId>
      <objectTypeName>Datastore</objectTypeName>
      <vsmUuid>42007DBE-3706-0598-4AEE-CCD932F94B3A</vsmUuid>
      <nodeId>1b060741-08e6-42a1-9dee-b39ea845337d</nodeId>
      <revision>1</revision>
      <type>
        <typeName>Datastore</typeName>
      </type>
      <name>iSCSI-3</name>
      <clientHandle></clientHandle>
      <extendedAttributes />
      <isUniversal>false</isUniversal>
      <universalRevision>0</universalRevision>
    </datastoreInfo>
    <controllerClusterStatus />
    <diskLatencyAlertDetected>FALSE</diskLatencyAlertDetected>
    <vmStatus>POWERED_ON</vmStatus>
  </controller>
</controllers>

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

<?xml version="1.0" encoding="utf-8"?>

<objectTypeName>Controller</objectTypeName>

<isUniversal>false</isUniversal>

<id>controller-1</id>

<status>RUNNING</status>

<upgradeStatus>NOT_STARTED</upgradeStatus>

<objectTypeName>VirtualMachine</objectTypeName>

<type>

<typeName>VirtualMachine</typeName>

</type>

<name>

NSX_Controller_78bdb508-aaea-41e3-aed9-1b10e296b323</name>

<scope>

<id>domain-c7</id>

<objectTypeName>ClusterComputeResource</objectTypeName>

<name>Compute Cluster-Site-A</name>

</scope>

<isUniversal>false</isUniversal>

</virtualMachineInfo>

<objectTypeName>HostSystem</objectTypeName>

<type>

<typeName>HostSystem</typeName>

</type>

<name>esxi-01a.corp.local</name>

<scope>

<id>domain-c7</id>

<objectTypeName>ClusterComputeResource</objectTypeName>

<name>Compute Cluster-Site-A</name>

</scope>

<isUniversal>false</isUniversal>

</hostInfo>

<objectId>resgroup-8</objectId>

<objectTypeName>ResourcePool</objectTypeName>

<type>

<typeName>ResourcePool</typeName>

</type>

<name>Resources</name>

<scope>

<id>domain-c7</id>

<objectTypeName>ClusterComputeResource</objectTypeName>

<name>Compute Cluster-Site-A</name>

</scope>

<isUniversal>false</isUniversal>

</resourcePoolInfo>

<objectId>domain-c7</objectId>

<objectTypeName>ClusterComputeResource</objectTypeName>

<type>

<typeName>ClusterComputeResource</typeName>

</type>

<name>Compute Cluster-Site-A</name>

<scope>

<id>datacenter-2</id>

<objectTypeName>Datacenter</objectTypeName>

</scope>

<isUniversal>false</isUniversal>

</clusterInfo>

<isDiskReadOnly>false</isDiskReadOnly>

<objectId>datastore-41</objectId>

<objectTypeName>Datastore</objectTypeName>

<type>

<typeName>Datastore</typeName>

</type>

<name>iSCSI-3</name>

<isUniversal>false</isUniversal>

</datastoreInfo>

<diskLatencyAlertDetected>FALSE</diskLatencyAlertDetected>

<vmStatus>POWERED_ON</vmStatus>

</controller>

</controllers>

Note: You can grep for keyword <id> in above example to limit the output of API call

Step-2: Delete NSX Controller

# Delete NSX Controller 
# CONTENT-TYPE: application/xml
# METHOD: DELETE
# API CALL: https://NSX-MGR-IP/api/2.0/vdn/controller/controller-xxx?forceRemoval=True

#EXAMPLE RESPONSE

200: OK

# Delete NSX Controller

# CONTENT-TYPE: application/xml

# METHOD: DELETE

# API CALL: https://NSX-MGR-IP/api/2.0/vdn/controller/controller-xxx?forceRemoval=True

#EXAMPLE RESPONSE

200: OK

Example: curl -sik -u “vcadmin@corp.local” -H ‘Content-Type: application/xml’ -X DELETE https://nsxmgr-01a.corp.local/api/2.0/vdn/controller/controller-1?forceRemoval=True

In vCenter server you will see a power off and delete task kicked for NSX controller VM. This task is kicked by the service account via which NSX is registered to vCenter server.

I hope you enjoyed reading this post. Feel free to share this on social media if it is worth sharing. Be sociable

Spread the Love

Leave a ReplyCancel reply