Source : Nutanix KB
Note: This KB applies to NCC 2.0. For NCC 2.3 or later, please see KB 3620
Note: NCC 2.0 is not supported on NOS versions other than 4.1.3.
The ncc_health_checks run_all command might fail after NCC upgrade.
After running the checks, you might see the following messages:
Old NCC process found on the Controller VMs, Please contact Nutanix support
Old NCC process with PID 5855,6075,13707,13879,17667,17879,19887,20108,20673,20898,22677,22867found on
Log on to any Controller VM in the cluster and restart the Prism web console service.
Note: The cluster start command restarts the Prism service only in this case, not the entire cluster.
nutanix@cvm$ allssh ‘~/cluster/bin/genesis stop prism’
nutanix@CVM$ cluster start
Note: In a recent case we needed to reboot each CVM. This was after trying the genesis stop prism and killing pids with pkill.
In case you see multiple PIDs in each CVMs:
Old NCC process found on the cvms, Please contact Nutanix support
Old NCC process with pid 465,466,467,468,481,503,504,19356 found on 10.81.29.23
Old NCC process with pid 8028,8050,8051 found on 10.81.29.26
Old NCC process with pid 7530,7552,7553 found on 10.81.29.29
Old NCC process with pid 30634,30656,30657 found on 10.81.29.32
Check if email_alert is configured and if so, disable it.
nutanix@cvm$ ncc –show_email_config
[ info ] NCC is set to send email every 12 hrs.
[ info ] NCC email has not been sent after last configuration.
[ info ] NCC email configuration was last set at 2015-08-24 21:29:44.243925.
nutanix@cvm$ ncc –delete_email_config