Segment created in NSX are in pending config state from manager view, with delayed creation in the vCenter
search cancel

Segment created in NSX are in pending config state from manager view, with delayed creation in the vCenter

book

Article ID: 323545

calendar_today

Updated On: 02-14-2025

Products

VMware NSX

Issue/Introduction

  • After creating segment on NSX-T (using either Policy, or Manager view), creation of segment is stuck in pending config-state.
  • After some time (typically 10-20 minutes), segment automatically gets into success config-state.
    Note: Time it takes to transition into success config-state has no pattern and it varies every time.
  • This can also impact creation of HCX service mesh.
  • You can observe segment creation request in NSX Manager's /var/log/policy/policy.log:

2023-06-13T09:23:12.739Z INFO http-nio-127.0.0.1-6440-exec-10901 FacadeInterceptorHelperImpl 6403 - [nsx@6876 comp="nsx-manager" level="INFO" subcomp="policy"] Starting intent for /policy/api/v1/infra/segments/<segment_name> with reqID 21894990-####-####-####-054dd6071a69

This is followed by creation of the segment:
2023-06-13T09:23:12.764Z INFO http-nio-127.0.0.1-6440-exec-10901 PolicyConnectivityFacadeImpl 6403 POLICY [nsx@6876 comp="nsx-manager" level="INFO" reqId="21894990-####-####-####-054dd6071a69" subcomp="policy" username="admin@example.com"] Segment with name /infra/segments/<segment_name> created/updated

Logical switch gets created:
2023-06-13T09:23:12.905Z INFO providerTaskExecutor-11 SegmentProviderNsxT 6457 POLICY [nsx@6876 comp="nsx-manager" level="INFO" subcomp="policy"] Creating the logical switch for segment /infra/segments/<segment_name> on NSX
..
2023-06-13T09:23:12.905Z INFO providerTaskExecutor-11 NsxTRestClient 6457 POLICY [nsx@6876 comp="nsx-manager" level="INFO" subcomp="policy"] NSX API POST http://127.0.0.1:7440/nsxapi/api/v1/logical-switches is called with LogicalSwitchDto{uplinkTeamingPolicyName='null', switchType='null', transportZoneId='f9755758-####-####-####-e7ccdd3a483b', vniPoolId='null', vni='null', vlan='null', vlan.cloud.corpnkSpec='null', adminState='UP', replicationMode='MTEP', addressBindings='null', switchingProfileIds='null', macPoolId='null', ipPoolId='null', hybrid='false', globalVni='null', span='null', extraConfigs='null', nsArpEnabled='null', recommended='null', super{ManagedResource{resourceType='null', aCreateUser='null', aCreateTime='null', aLastModifiedUser='null', aLastModifiedTime='null', aSystemOwned='null', aProtection='null', id='0bda9243-####-####-####-94ca360ab38d', displayName='<segment_name>', description='null', tags='[Tag{scope='policyPath', tag='/infra/segments/<segment_name>'}, Tag{scope='subnet-cidr', tag='#.#.#.#/24'}]', super{RevisionedResource{aRevision='null', super{Resource{aSelf='null', aLinks='null', aSchema='null'}}}}}}} and
..
2023-06-13T09:23:13.059Z INFO providerTaskExecutor-11 SegmentProviderNsxT 6457 POLICY [nsx@6876 comp="nsx-manager" level="INFO" subcomp="policy"] Created the logical switch for segment /infra/segments/<segment_name> on NSX

After this, logical switch is added to to the fabric:
/var/log/syslog:
2023-06-13T09:24:05.122Z INFO LogicalSwitchStateOperationScheduler2 LogicalSwitchStateAutoRectifier 6063 SWITCHING [nsx@6876 comp="nsx-manager" level="INFO" subcomp="manager"] LogicalSwitchStateAutoRectifier: registering out-of-sync-hosts for LogicalSwitches: [LogicalSwitch/548ff647-####-####-####-ac4a41b936ff, LogicalSwitch/0bda9243-####-####-####-94ca360ab38d]

Number of logical switches stuck in in-progress stage, e.g.:
/var/log/syslog:
2023-06-13T13:48:45.605+03:00 pd10-nmg-02.vlprs.local NSX 6084 SWITCHING [nsx@6876 comp="nsx-manager" level="INFO" subcomp="manager"] LSAutoRectifier: LS 6abe9d0f-####-####-####-99ed61e94109 is in IN_PROGRESS state and out of sync host list in empty. Adding request to sync.

Some time (typically 10 to 20 minutes) later, the system will successfully realize the segment:
/var/log/proton/proton/nsxapi.log 
2023-06-13T09:34:30.906Z  INFO LogicalSwitchStateOperationScheduler1 NsxPortgroupExecuteVcOperationImpl 6063 SWITCHING [nsx@6876 comp="nsx-manager" level="INFO" subcomp="manager"] Cm request for creating lspg. cmId : 2a6d4e72-####-####-####-841c46e40aa3, vdsId : pd10-###-#############, isResync: false, portGroupObjDto : NsxPortgroupObjDto{cmId='2a6d4e72-####-####-####-841c46e40aa3', vdsMoref='dvs-######', nsxPortgroups='[NsxPortgroupDto{nsxPortgroupMoref='null', nsxPortgroupKey='null', nsxPortgroupName='<segment_name>', nsxPortgroupType='ephemeral', configVersion='null', numPorts='null', backingType='nsx', vlanId='null', vni='#####', transportZoneUuid='f9755758-####-####-####-e7ccdd3a483b', transportZoneName='pd10-##########', logicalSwitchUuid='0bda9243-####-####-####-94ca360ab38d', segmentId='/infra/segments/<segment_name>'}]', super{CmRequestParameter{parameterType='NsxPortgroupObj', parameterDetails='null', ignoreExceptionType='null'}}}

 

Environment

VMware NSX-T Data Center
VMware NSX

Cause

Logical switch operation is in progress due to logical switches that are already deleted being stuck in in-progress state, which will cause a delay in segment creation.

Resolution

This issue is resolved in VMware NSX 4.1.2
This issue is resolved in VMware NSX 4.2.0

Workaround:

In older versions than the fixed builds, if the realization status is not updated after waiting the indicated typical delay duration (10 to 20 minutes) you may attempt a rolling reboot of all the Manager nodes of the NSX Manager cluster in order to force the autorealization workflow into resuming.

Powered by Wolken Software