Oracle cold Backup Recovery Nospaceleftondevice

Environment OS: centOS6.3oracle: Version11.2.0.1.0 enterprise needs to restore Directory: $ ORACLE_BASEoradatamd or u01apporacleoradatamd cold Backup Directory: tmpcoldbackup error description during oracle cold Backup Recovery (the database is closed ), copy the cold backup data file to the directory to be restored

Environment OS: centOS 6.3 oracle: Version 11.2.0.1.0 enterprise needs to restore the Directory: $ ORACLE_BASE/oradata/md/or/u01/app/oracle/oradata/md cold Backup directory: /tmp/coldbackup/error description when performing oracle cold Backup Recovery (the database is closed), copy the cold backup data file to the directory to be restored

Environment
OS: centOS 6.3

Oracle: Version 11.2.0.1.0 Enterprise Edition

Directory to be restored: $ ORACLE_BASE/oradata/md/or/u01/app/oracle/oradata/md

Cold Backup Directory:/tmp/coldbackup/

Error description

When performing oracle cold Backup Recovery (the database is closed), when copying the cold backup data file to the directory to be restored, the linux system reports the following error: No space left on device.

Operations and Solutions

In the above operation, we can see that the space is still 1.7G, and 51% is used.

The operation above shows that there are no files in $ ORACLE_BASE/oradata/md. The reason is that the space usage is 0%.

Check whether inode in the $ ORACLE_BASE/oradata/md directory is exhausted. We can see that 20 or 231052 of inode are idle,

Http://www.zeali.net/entry/480)

The above shows that there are still processes in linux occupying deleted files. Although ls-la cannot find files on the disk, it actually occupies space, it can be seen that process 6576 occupies resources.

It can be seen that there is indeed a process 6576.Here, I used kill-9 6576 to kill the process, and the problem was solved. I used df to check that the space used in the past reached 51%, for example:

Reference (http://www.cnblogs.com/Richardzhu/p/3142076.html) for this issue)