A, description,
An Oracle11g database cluster environment needs to be set up. During disk mapping, the fdisk command on Linux is used to mount disk partitions. However, an error message is displayed after the partitions are mapped. You need to run the partprobe command to view the partition list. After one node is restarted, the zone is lost. After two nodes are restarted, the zone is not lost (using the zone of one node).
Second, the problem of
After partition alarm is as follows:
WARNING: re-reading the partition table failed with error 16: The device or resource is busy. The kernel still uses the old table. The new table will be used at the next reboot or after you run partprobe(8) or Kpartx (8) is synchronizing disks.
After a partition is created, the partition list cannot be viewed in /dev. You need to run partProbe to view the partition list. Node 2 also needs to execute to see the partition list.
After the restart, the partition list is lost
The sda disk is not in the /etc/fstab directory.
Third, solutions
Umount the disk once before partitioning.
# umount /dev/sda
Then partition the disk. Check that the partition is correct.
4. Restart verification
After the server is restarted, the disk partition is not lost. Problem solved.
One more thing:
This problem was also troubled for a while, because there was no sDA mounting conflict found in the /etc/fstab file, so I did not think of such a solution, and the online solution was not found. With a try mentality, the problem was solved. The new man white, record the small problems encountered daily. If you have the same problem as me, I hope it can help you.