Title: vCenter Server ELM Re-join/Repoint Fails with Error: "Already exists / ERROR_TOO_MANY_NAMES (68)" During vdcpromo
search cancel

Title: vCenter Server ELM Re-join/Repoint Fails with Error: "Already exists / ERROR_TOO_MANY_NAMES (68)" During vdcpromo

book

Article ID: 452966

calendar_today

Updated On:

Products

VMware vCenter Server

Issue/Introduction

  • Deploying or re-joining a replacement vCenter Server Appliance into an existing Enhanced Linked Mode (ELM) domain fails during the vdcpromo stage.

  • Deployment logs or /var/log/firstboot/vmdir-firstboot.py_*.log throw an LDAP error:

    Already exists / ERROR_TOO_MANY_NAMES (68)

  • The environment contains a standalone vCenter Serverthat needs to be re-integrated into the existing ELM domain, but stale entries for a previously failed node remain in the directory.

 

Environment

vCenter Server 8.x

Cause

During an aborted deployment or failed upgrade attempt, metadata objects, solution users, and Service Principal Names (SPNs) for the node are partially written to the VMware Directory Service (vmdir) database. Because the deployment terminates abruptly, automated cleanup tasks never run, leaving orphaned node records. Subsequent join or repoint attempts fail because the directory service detects these existing duplicate entries, triggering the ERROR_TOO_MANY_NAMES (68) schema constraint error.

Resolution

  1. Take offline (powered-off) snapshots of all healthy, surviving vCenter Server nodes in the existing ELM domain, as well as the standalone

  2. SSH into a surviving, healthy vCenter Server in the existing SSO domain.

  3. Verify the stale node registration using the vdcrepadmin utility: /usr/lib/vmware-vmdir/bin/vdcrepadmin -f showservers -h localhost -u administrator

  4. Unregister the stale vCenter Server entry: cmsso-util unregister --node-pnid <OLD_VC_FQDN>  - -username [email protected]

  5. Perform a pre-check on clark.local to evaluate potential domain repointing conflicts: cmsso-util domain-repoint -m pre-check --src-emb-admin Administrator --replication-partner-fqdn <Existing_vCenter_FQDN> --replication-partner-admin Administrator --dest-domain-name <Existing_SSO_Domain>

  6. Execute the repointing command to reintroduce clark.local into the existing ELM topology: cmsso-util domain-repoint -m execute --src-emb-admin Administrator --replication-partner-fqdn <Existing_vCenter_FQDN> --replication-partner-admin Administrator --dest-domain-name <Existing_SSO_Domain>

Additional Information

Refer the above KB

https://knowledge.broadcom.com/external/article/450887/repointing-vcenter-server-to-a-new-sso.html

Powered by Wolken Software