Hi Stefan,
On Mon, Jul 18, 2011 at 10:43 AM, Stefan Priebe - Profihost AG <s.pri...@profihost.ag> wrote: > > thanks for fixing the novell page. It really confused me ;-) You're welcome. Thanks for pointing it out. > >> This was an error and we have fixed the document. However your problem >> could be caused due to different reasons. Most often it is >> fragmentation. An output of >> http://oss.oracle.com/~smushran/debug/scripts/stat_sysdir.sh may help >> us identify the cause. > > Output is attached. > Your filesystem does look fragmented, and the the inode_alloc (002) is running low. However, the filesystem is not running out of inodes or extent_blocks, unless you deleted some files. I think you would be able to find out something after enabling some logs using debugfs.ocfs2 and re-creating the situation. I suggest you start with - DISK_ALLOC AIO FILE_IO NAMEI ENTRY EXIT CONFIG_OCFS2_DEBUG_MASKLOG needs to be set in the kernel. HTH, -- Goldwyn _______________________________________________ Ocfs2-users mailing list Ocfs2-users@oss.oracle.com http://oss.oracle.com/mailman/listinfo/ocfs2-users
