Messages in this thread | | | Date | Sun, 10 Jan 1999 17:25:08 +0000 (GMT) | From | Matthew Kirkwood <> | Subject | Re: ext2 corruption? |
| |
On Sat, 9 Jan 1999, Goran Gajic wrote:
> galeb-Sat09-12:38pm/users/etf/chuka/.tin# ls -al > total 18 > drwx------ 5 chuka etf 1024 May 28 1998 ./ > drwx--x--x 9 chuka etf 1024 Jan 2 19:47 ../ > drwx------ 2 chuka etf 1024 Apr 26 1996 .mail/ > drwx------ 2 chuka etf 1024 Apr 26 1996 .news/ > drwx------ 2 chuka etf 1024 Apr 26 1996 .save/ > -rw------- 1 chuka etf 100664896 Apr 26 1996 attributes > -rw------- 1 chuka etf 67110035 Nov 6 1997 filter > -rw------- 1 chuka etf 67108951 Apr 26 1996 headers > -rw------- 1 chuka etf 420 May 28 1998 newsgroups > -rw------- 1 chuka etf 67108941 Apr 26 1996 posted > -rw------- 1 chuka etf 7430 May 28 1998 tinrc > galeb-Sat09-12:38pm/users/etf/chuka/.tin# du -sk > 17 . > galeb-Sat09-12:38pm/users/etf/chuka/.tin# > > Kernel is 2.0.36. I have no idea how/when did this happen. e2fsck > doesn't find anything wrong (e2fsprograms version is 1.12). Btw. I have > expirienced this problem only on partitions that were mke2fs long time > ago..
Your files probably have holes in them. Example:
1. Write a character at position 1 2. Seek +10M 3. Write another character.
The file is now 10M+2 long, but only two blocks have been allocated. read()s will return zero, and you'll be able to read 10M+2 bytes from the file, but stat() (and the stat program which ships with many distributions) will say size=10M+2, blocks=2 (actually more for indirect blocks).
tin is probably playing similar tricks.
Summary: everything is fine, don't worry.
Matthew.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |