After the active and standby N2000 BMS servers are installed, restart them and find that the /opt/n2000 and /opt/sybase directories are empty. The database cannot be started. The watchman can be started properly but the replication relation of the watchman turns orange. In this case, the replication relation cannot be rebuilt after it is deleted.
1. Modify the names of disk arrays for the active and standby N2000 BMS servers. The names of disk arrays for the active and standby N2000 BMS servers must be the same, that is, n2kdg_host name of the active server. The modification method is as follows:
IP address: 127.0.0.1
User name: root
Password: password of the root user
On the GUI, choose the disk array and change its name to that of n2kdg_N2000Primary. At the same time, modify the /etc/vfstab file manually. Change the name of the disk array to that of n2kdg_N2000Primary (do not delete the "#" preceding the name).
2. Add the information of the replication volume to the watchman client manually. For details, see the help of watchman.
3. After the modification is finished, start the active N2000 BMS server. The problem is solved.
1. Use the collection tool of the watchman to collect information on the active and standby servers at the same time.
2. Run the following commands:
mount /dev/vx/dsk/n2kdg_N2000Primary/lv_sybase ?/opt/sybase
mount /dev/vx/dsk/n2kdg_N2000Primary/lv_n2000 /opt/n2000
The contents of the cat /etc/vfstab file are returned.
3. The /opt/watchman/server/conf/watchman.xml file shows that the related fields are empty on the two servers. It indicates that the installation of the HA system is abnormal. That is, the HA system does not perform any operation on the lv_n2000 and lv_sybase volumes that corresponds to /opt/n2000 and opt/sybase respectively.
4. The duplication disk groups of the active and the standby servers are different, that is, “n2kdg_N2000Primary” on the active server, and “n2kdg_N2000Secondary” on the standby server.