I was trying to add some node in our existing 6 node clusters, out ./cluvfy stage -pre crsinst was failing because of NTP sync error .
[grid@exa01 bin]$ ./cluvfy stage -pre crsinst -n exa05 -verbose Check for NTP daemon or service using UDP port 123 Node Name Port Open? ------------------ ---------- exa05 yes PRVG-13602 : NTP daemon is not synchronized with any external time source on node " exa05 ". Result: Clock synchronization check using Network Time Protocol(NTP) failed
I tried to check if the ntp server conf file was missing the ntp server name
I checked the /etc/ntp.conf file , but the file was looking good.
[root@exa05 ~]# grep ^server /etc/ntp.conf server 10.44.26.21 prefer iburst burst minpoll 4 maxpoll 4 server 10.44.26.51 iburst burst minpoll 4 maxpoll 4
I try to run clocksync command as well , to get any more information
[grid@exa05 bin]$ ./cluvfy comp clocksync Verifying Clock Synchronization across the cluster nodes Oracle Clusterware is installed on all nodes. CTSS resource check passed Query of CTSS for time offset passed CTSS is in Observer state. Switching over to clock synchronization checks using NTP Starting Clock synchronization checks using Network Time Protocol(NTP)... NTP configuration file "/etc/ntp.conf" existence check passed Liveness check passed for "ntpd" Check for NTP daemon or service alive passed on all nodes PRVG-13602 : NTP daemon is not synchronized with any external time source on node "oralux401". Clock synchronization check using Network Time Protocol(NTP) failed PRVF-9652 : Cluster Time Synchronization Services check failed
Still it failed
When NTP is configured , it is to be ensured that NTP configuration points to an external time reference to which the nodes are kept in sync.Once NTP is disabled, CTSS will become active and maintain time in sync across cluster nodes. However, cluster time may get out of sync with the wall clock time because CTSS also does not sync with a reference time.
For the workaround , we disabled the NTP and made CTSSD as active time sync tool from its usual observer mode
Disable the NTP
1. service ntp stop [root@exa01 ~]# service ntpd stop Shutting down ntpd: [ OK ] 2. mv /etc/ntp.conf /etc/ntp.conf_bkp 3. service ntpd status
After disabling NTP the cluvfy precheck passed and addnode was successful
Later again , we restoted the NTP conf.
1.mv /etc/ntp.conf_bkp /etc/ntp.conf 2.service ntp start [root@exa01 ~]# service ntpd start Starting tpd: [ OK ] 3.service ntpd status
553 total views, 4 views today