Description
The system checks its time every minute. This alarm is generated when the system detects that the time offset exceeds 5 minutes.
Impact on the system
Services on a node are affected.
The alarm generated during the time offset or adjustment period must be manually cleared.
Possible Causes
1. The administrator modified the system time.
2. The administrator modified the system time zone.
Procedure
Handle the alarm.
1. In the navigation tree on the left choose .
2. On the Real-Time Alarms tab, locate the row that contains the target alarm and click the name in the Alarm Object column.
3. The page is displayed.
4. On the Services and Management Nodes page, view the management node information and obtain the management IP address of the Alarm Object viewed in 3.
5. Click System Configuration on the System Management page, select and obtain the IP address of the NTP Server.
6. Use PuTTY to log in to the node for which the alarm is generated.
7. Ensure that the management IP address and username gandalf are used to establish the connection.
8. The system supports the login authentication using a password or private-public key pair. If you use a private-public key pair to authenticate the login, see Using PuTTY to Log In to a Node in Key Pair Authentication Mode.
9. Run the following command and enter the password of user root to switch to user root: su - root.
10. Check whether the NTP Server is configured.
a) If yes, go to 9.
b) If no, go to 12.
11. Have the administrator check whether the node can properly communicate with the NTP server and whether the NTP server runs properly. After the heartbeat communication between the node and the NTP server is restored, go to 10.
12. Run the service ntpd stop;/usr/sbin/ntpdate NTP server IP address or domain name&& /sbin/hwclock -w -u;service ntpd start command to synchronize time.
NOTE:
If the command output contains the NTP socket is in use, exiting, run this command again.
13. Check whether no server suitable for synchronization found is contained in the command output.
a) If yes, go to 15.
b) If no, go to 12.
14. Run the following command to restart all processes on the node: sh /opt/galax/gms/common/config/restartProcess.sh.
NOTE:
The restart of all processes takes about 3 minutes. During the restart, you cannot log in to the FusionCompute portal.
15. Check whether the following information is displayed: restart all process success.
a) If yes, go to 14.
b) If no, go to 15.
16. After 3 minutes, log in to FusionCompute again, locate the row that contains the to be cleared, click Clear, and clear this. Check whether the row is cleared. If yes, no further action is required.
Handle the impact.
The generation and processing involve time adjustment. If the time is adjusted to past time, some cannot be cleared automatically. For example, if the time is adjusted from T1 09:00 a.m. to T2 08:00 a.m., some times generated from T0 08:00 a.m. (original 08:00 a.m.) to T1 cannot be cleared automatically. These need to be cleared manually. After the time reaches T3 09:00 a.m., check whether the generated from T2 to T3 on the Monitoring page are automatically cleared based on the online help information.
If not, choose to manually clear the alarm.
