how to deal with Standby Node Database Is Abnormal alarm on FusionCompute V1R5C00 version

5

This alarm is generated when the system detects that the database on the standby Virtualization Resource Management (VRM) node does not work properly.

This alarm is cleared when the standby VRM node database works properly.
On the FusionCompute portal, click System, and choose System Configuration > Service and Mgmt. Node in the navigation tree on the left.

On the displayed Service and Mgmt. Node page, obtain the standby VRM node name in the table of the Service list area, and view the IP address of the standby VRM node in the table of the Management node area.


Use PuTTY to log in to the standby VRM node.
Ensure that the management IP address and username gandalf are used to establish the connection.

Run the following command and enter the password of user root to switch to user root:

su - root


Run the following command to stop the High Availability (HA) process on the standby node:
service had stop

Run the following command to stop the database process:
su - postgres -c "gs_ctl stop"

Run the following command to start the database process:
su - postgres -c "gs_ctl start -M standby"

Check whether the database is started:

If yes, go to Step 8.
If no, go to Step 12.

Run the following command to rebuild the database:
su - postgres -c "gs_ctl build -b incremental -M standby"

Check whether the database is rebuilt:

If yes, go to Step 10.
If no, go to Step 12.

Run the following command to start the HA process:
service had start

After about 5 minutes, log in to the FusionCompute portal and check whether the alarm is cleared.

Other related questions:
how to deal with PMCD Process Is Abnormal alarm on FusionCompute V1R5C00 version
This alarm is generated when the system detects that the monitoring processing process pmcd has stopped or is faulty. This alarm is cleared when the pmcd process becomes normal. Identify the management IP address of the node for which the alarm is generated. On the FusionCompute portal, click Monitoring, and choose Alarm in the navigation tree on the left. On the Alarm page, locate the row that contains the alarm and click the name in the Alarm Object column. The Service and Management Node page is displayed. On the Service and Management Node page, view information about the management nodes and obtain the IP address of the node whose name is the same as that of Alarm Object in Step 1. Restart the process of the performance management center. Use PuTTY to log in to the node. Ensure that the management IP address and username gandalf are used to establish the connection. Run the following command and enter the password of user root to switch to user root: su - root Run the following command to restart the process: service pmcd restart NOTE: Restarting the process may cause temporary interruption of monitoring processing service. Run the following command to check the process status: service pmcd status Check whether the command output is as follows: Checking for pmcd: running If yes, go to Step 8. If no, go to Step 9. After 3 or 4 minutes, switch to the Alarm page and check whether the alarm is cleared.

how to deal with Performance Monitoring Process on Management Nodes Is Abnormal alarm on FusionCompute V1R5C00 version
This alarm is generated when the system detects that the performance monitoring process on management nodes has stopped or is faulty. This alarm is cleared when the performance monitoring process becomes normal. Identify the IP address of the object for which the alarm is generated. On the FusionCompute portal, click Monitoring, and choose Alarm in the navigation tree on the left. On the Alarm page, locate the row that contains the alarm and click the name in the Alarm Object column. The Service and Management Node page is displayed. On the Service and Management Node page, view information about the management nodes and obtain the IP address of the node whose name is the same as that of Alarm Object in Step 1. Restart the performance monitoring process on management nodes. Use PuTTY to log in to the node. Ensure that the management IP address and username gandalf are used to establish the connection. Run the following command and enter the password of user root to switch to user root: su - root Run the following command to restart the process: service pmad restart NOTE: Restarting the process may cause temporary interruption of the performance monitoring service. Run the following command to check the process status: service pmad status Check whether the command output is as follows: Checking for service pma running If yes, go to Step 8. If no, go to Step 9. After 3 or 4 minutes, switch to the Alarm page and check whether the alarm is cleared.

how to deal with NOTIFY Process Is Abnormal alarm on FusionCompute V1R5C00 version
This alarm is generated when the system detects that the subscription notification process has stopped or is abnormal. This alarm is cleared when the subscription notification process becomes normal. Identify the management IP address of the Virtualization Resource Management (VRM) node. On the FusionCompute portal, click Monitoring, and choose Alarm in the navigation tree on the left. On the Alarm page, locate the row that contains the alarm and click the name in the Alarm Object column. The Service and Management Node page is displayed. On the Service and Management Node page, view information about the management nodes and obtain the IP address of the node whose name is the same as that of Alarm Object in Step 1. Restart the subscription notification process. Use PuTTY to log in to the node. Ensure that the management IP address and username gandalf are used to establish the connection. Run the following command and enter the password of user root to switch to user root: su - root Run the following command to restart the process: service notifyd restart NOTE: Restarting the process may cause temporary interruption of the subscription notification service. Run the following command to check the process status: service notifyd status Check whether the command output is as follows: Checking for notifyd: running If yes, go to Step 8. If no, go to Step 9. After 3 or 4 minutes, switch to the Alarm page and check whether the alarm is cleared.

how to deal with VNC Agent Process Is Abnormal alarm on FusionCompute V1R5C00 version
This alarm is generated when the system detects that the VNC agent process vncd has stopped or is faulty. This alarm is cleared when the vncd process becomes normal. Identify the management IP address of the node for which the alarm is generated. On the FusionCompute portal, click Monitoring, and choose Alarm in the navigation tree on the left. On the Alarm page, locate the row that contains the alarm and click the name in the Alarm Object column. The Service and Management Node page is displayed. On the Service and Management Node page, view information about the management nodes and obtain the IP address of the node whose name is the same as that of Alarm Object in Step 1. Restart the process of the performance management center. Use PuTTY to log in to the node for which the alarm is generated. Ensure that the management IP address and username gandalf are used to establish the connection. Run the following command and enter the password of user root to switch to user root: su - root Run the following command to restart the process: service vncd restart NOTE: Restarting the process may cause temporary interruption of VNC agent processing service. Run the following command to check the process status: service vncd status Check whether the command output is as follows: Checking for vncd: running If yes, go to Step 8. If no, go to Step 9. After 3 or 4 minutes, switch to the Alarm page and check whether the alarm is cleared

how to deal with Database Space Is Going to Be Insufficient alarm on FusionCompute V1R5C00 version
This alarm is generated when the system detects that the database space usage of management nodes hits 80%. This alarm is cleared when the database space usage is less than 80%. Confirm the IP address of the faulty node. On the FusionCompute portal, click Monitoring, and choose Alarm in the navigation tree on the left. On the Alarm page, locate the row that contains the alarm and click the name in the Alarm Object column. The Service and Management Node page is displayed. On the Service and Management Node page, view information about the management nodes and obtain the management IP address of the node whose name is the same as that of Alarm Object in Step 1. Rectify the fault. Use PuTTY to log in to the faulty node. Ensure that the management IP address and username gandalf are used to establish the connection. Run the following command and enter the password of user root to switch to user root: su - root Run the following command to query the space usage of each database. df -l Information similar to the following is displayed: VRM8810:~ # df -l Filesystem 1K-blocks Used Available Use% Mounted on /dev/sda1 10317828 2033244 7760468 21% / devtmpfs 8147620 212 8147408 1% /dev tmpfs 8147620 0 8147620 0% /dev/shm /dev/sda6 1027768 17912 957648 2% /etc/galax /dev/sda10 880874308 204948 835923520 1% /extend /dev/sda8 1106800 34188 1016388 4% /opt/galax/upgrade /dev/mapper/vg_vrm-lv_gaussdb 18578172 627648 17006808 4% /opt/gaussdb /dev/sda5 18572112 316568 17312132 2% /var /dev/mapper/vg_vrm-lv_backup 30834692 176196 29092188 1% /var/backup NOTE: Filesystem: specifies the name of a disk partition. 1K-blocks: specifies the disk capacity. Used: specifies the used disk capacity. Available: specifies the available disk capacity. Use%: specifies the disk usage. Mounted on: specifies the directory to which the disk is attached. Enter the directory that contains the alarm object, and delete the files that are unnecessary or that you have copied. NOTICE: File deletion is a high-risk operation. Therefore, make sure that the files to be deleted are not those that are stored in the system directory or in the directory that contains important in-use services. After 10 or 15 minutes, check whether the alarm is cleared

If you have more questions, you can seek help from following ways:
To iKnow To Live Chat
Scroll to top