From: Taylor, David (DTaylor@WBMI.COM)
Date: Mon May 19 2003 - 00:51:57 EDT
I have a large (800 GB) file-system that is used as cache in front of an optical jukebox. The server is AIX 5.1 ML3. The disk is fibre attached Hitachi storage. The files are small ~100K images with a 200 byte index file. This file-system was about half full. It became corrupted. I ran fsck and after about 45 minutes it began giving errors stating that it was unable to expand lost+found and there were no available direct blocks. After fsck finished, I remounted the file-system and noted that I still had several hundred GB available and only 19% of the inodes were used. The file-system is mounted with the "mind" option.
What caused this (the issue with lost+found)?
Is there a limitation on the number of files that can be contained within a directory?
I know what a direct-inode is. What is a direct block?
IBM stated that the only option was to move everything out of lost+found and rerun fsck - or blow away the whole file-system and rebuild from tape - did I mention that we use TSM (no colocated) and the file-system has taken 2 years to get to its current size!
I still am trying to discover how it became corrupt. I had no error entries until after running fsck. The only reason I ran fsck was I noticed that an empty directory disappeared and when I attempted to recreate it I got really weird errors.
TIA
David
**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
This footnote also confirms that this email message has been swept by
MIMEsweeper for the presence of computer viruses.
www.mimesweeper.com
**********************************************************************
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 22:16:50 EDT