Ran into an issue last week were a 150GB OCFS2 partition would not allow any further data to be written to it. The issue was first discovered as a result of a PHP web app alarming that it couldn't write files to disk. When the issue was investigated initial indications showed; via `df -h`, that the partition still had 56GB of space left. When an an attempt to create a random file using dd was performed it ran into the same issue.
Realizing that the server is acting as a file server and that a PHP web app is processing imports and exports of small files it started looking like the OCFS2 fs was fragmented which resulted in unusable disk space that appears via `df` to be available. We have since moved to a much larger partition but realize that without knowing the fragmentation "ceiling" that exists with regards to writing/deleting small files on this partition we will eventually run into the same thing. First thought was, how can the OCFS2 FS be defragmented, however I realize that there might also be some tuning that could be done to minimize fragmentation in the first place. Can anyone help me with either of these concepts? Thanks! David _______________________________________________ Ocfs2-users mailing list Ocfs2-users@oss.oracle.com http://oss.oracle.com/mailman/listinfo/ocfs2-users
