Wednesday, February 2, 2022

Oracle 19c GI installation failed with PRVG-13605 : Query of chrony daemon for its time sources could not be performed

The issue happens on Oracle Linux / Red Hat Enterprise Linux 7/8 on which chronyd is adopted as the network time protocol daemon instead of ntpd. When installing Oracle Grid Infrastructure 19c on these platform, the installation fails during prerequisit check, the message shows that

Verifying Network Time Protocol (NTP) ...FAILED
  Verifying chrony daemon is synchronized with at least one external time
  source ...FAILED
  rac01: PRVG-13605 : Query of chrony daemon for its time sources could not
         be performed on node "rac01".
         PRVG-2043 : Command "/usr/bin/chronyc sources " failed on node
         "rac01" and produced the following output:
         506 Cannot talk to daemon

  rac02: PRVG-13605 : Query of chrony daemon for its time sources could not
            be performed on node "rac02".
            PRVG-2043 : Command "/usr/bin/chronyc sources " failed on node
            "rac02" and produced the following output:
            506 Cannot talk to daemon

Manually test the status of service chronyd,
[root@rac01]# systemctl status chronydchronyd.service - NTP client/server
   Loaded: loaded (/usr/lib/systemd/system/chronyd.service; enabled; vendor preset: enabled)
   Active: active (running) since Wed 2022-02-02 17:10:33 EST; 22s ago
     Docs: man:chronyd(8)
           man:chrony.conf(5)
  Process: 43769 ExecStartPost=/usr/libexec/chrony-helper update-daemon (code=exited, status=0/SUCCESS)
  Process: 43764 ExecStart=/usr/sbin/chronyd $OPTIONS (code=exited, status=0/SUCCESS)
 Main PID: 43768 (chronyd)
    Tasks: 1 (limit: 3355442)
   Memory: 376.0K
   CGroup: /system.slice/chronyd.service
           └─43768 /usr/sbin/chronyd

[root@rac01 ~]# chronyc sources
210 Number of sources = 4
MS Name/IP address         Stratum Poll Reach LastRx Last sample
===============================================================================
^? ts1.lab.dbaplus.ca            2   6     3    18   -166us[ -166us] +/-   17ms
^? ts2.lab.dbaplus.ca            2   6     3    17   +427us[ +427us] +/-   25ms

[root@rac01 ~]# su - grid
Last login: Wed Feb  2 20:11:48 EST 2022 on pts/1

[grid@rac01]$ chronyc sources
506 Cannot talk to daemon
The service "chronyd" is running and user root can run command "chronyc sources" successfully, and the command list all the time sources that chronyd is accessing. The sources are servers ts1.lab.dbaplus.ca and ts2.lab.dbaplus.ca.

However Grid Infrastructure (GI) owner grid cannot run the command, failed with message "506 Cannot talk to daemon".

The reason is that the system is stopping chrony daemon from listening on the UDP port 323 for management connections made by chronyc. This option is set by adding following line in chrony configuration file "/etc/chrony.conf"
cmdport 0
This line can be commented out (add # at the beginging of the line) to enable chronyc management connctions. The line as following
# cmdport 0
After saving the file "/etc/chrony.conf", restart chronyd service
systemctl restart chronyd
Now, normal user (non root) grid can run "chronyc sources" successfully.

No comments: