0 Replies Latest reply on Jul 15, 2020 6:38 AM by sc_2111

    decomission vcenter in ELM

    sc_2111 Enthusiast

      We have to decomission a vcenter with embedded PSC from an ELM configuration

      The main Vcenter with its embedded PSC will remain in HQ site .

      Following the kb VMware Knowledge Base  we performed the below steps


      On the vcenter/PSC to be decomissioned we first run the command to unregister the vcenter

      cmsso-util unregister --node-pnid prdvcsadr.xxx.com --username administrator@vsphere.local --passwd xxxx


      the command reported

      020-07-15T12:58:55.651Z   Running command: ['/usr/lib/vmware-vmafd/bin/dir-cli', 'service', 'list', '--login', 'administrator@vsphere.local']

      2020-07-15T12:58:55.754Z   Done running command

      2020-07-15T12:58:56.124Z   RC = 1

      Stdout = vdcleavefd offline for server prdvcsadr.xxx.com

      Leave federation cleanup failed. Error[1] - Operations error

      Stderr = password:

      ERROR unregistering computer account.


      2020-07-15T12:58:56.125Z   Running command: ['/usr/bin/sed', '-i', '-e', 's/cmsso-util.*/cmsso-util/g', '/var/log/vmware/procstate']

      2020-07-15T12:58:59.972Z   Done running command


      After that we checked the main Vcenter and we found that the remote vcenter was correctly removed


      So we went through the step to unregister the remote's PSC from the main site

      We shut down the DR appliance ( PSC) and performed the following command on the main site's appliance

      root@prdvcsa [ ~ ]# cmsso-util unregister --node-pnid prdvcsadr.xxxx.com - -username administrator@vsphere.local--passwd xxxx

      and got the following error


      Could not find a host id which maps to prdvcsadr.xxxx.com  in Component Manage r


      2020-07-15T13:22:17.609Z   Running command: ['/usr/bin/sed', '-i', '-e', 's/cmss o-util.*/cmsso-util/g', '/var/log/vmware/procstate']

      2020-07-15T13:22:17.709Z   Done running command


      Can anyone explain if this expected since we already unregister the DR's vcenter or if there's something wrong in the procedure ?