You write:
> Ok, this is probabally old news and has been fixed, but
> the following happened in kernel 2.4.3 (ironically when i was deleting
> /usr/src/linux in order to extract the latest 2.4.5 :-)
Old news for me, fixed this bug in January, see:
http://marc.theaimsgroup.com/?l=linux-kernel&m=98339241223296&w=4
but the whole patch did not make it into the Linus kernel (not sure
if it is in -ac or not).
> However, it has a very slight sound of a more serious problem, to do with my
> disk (however, since i know nothing about this error message, i will take no
> sides)
However, there is still a problem in the ext2 code because it does not
clear the in-memory inode on the error paths. Patch below will fix.
Linus (and Alan, if needed), please apply.
> May 29 01:36:35 toweringmeep kernel: hda: read_intr: status=0x59 {
> DriveReady SeekComplete DataRequest Error }
> May 29 01:36:35 toweringmeep kernel: hda: read_intr: error=0x40 {
> UncorrectableError }, LBAsect=2113616, sector=2113553
> May 29 01:36:35 toweringmeep kernel: end_request: I/O error, dev 03:01
> (hda), sector 2113553
> May 29 01:36:35 toweringmeep kernel: EXT2-fs error (device ide0(3,1)):
> read_inode_bitmap: Cannot read inode bitmap - block_group = 129,
> inode_bitmap = 1056776
> May 29 01:36:35 toweringmeep kernel: kernel BUG at inode.c:886!
> is this a nice catch 22 ? or is there a nasty problem with one of my disks
> ... im about to reboot and check the disk....
You have a disk problem, which caused I/O failure (the hda: read_intr
messages are first). Time for a backup (if you don't have one) and a
new disk.
Cheers, Andreas
===========================================================================
--- linux-2.4.5.orig/fs/ext2/ialloc.c Tue Apr 10 16:44:49 2001
+++ linux-2.4.5-aed/fs/ext2/ialloc.c Tue May 29 13:32:04 2001
@@ -199,10 +199,15 @@
lock_super (sb);
es = sb->u.ext2_sb.s_es;
- if (ino < EXT2_FIRST_INO(sb) ||
+ is_directory = S_ISDIR(inode->i_mode);
+
+ /* Do this BEFORE marking the inode not in use or returning an error */
+ clear_inode (inode);
+
+ if (ino < EXT2_FIRST_INO(sb) ||
ino > le32_to_cpu(es->s_inodes_count)) {
- ext2_error (sb, "free_inode",
- "reserved inode or nonexistent inode");
+ ext2_error (sb, "ext2_free_inode",
+ "reserved or nonexistent inode %lu", ino);
goto error_return;
}
block_group = (ino - 1) / EXT2_INODES_PER_GROUP(sb);
@@ -210,13 +215,8 @@
bitmap_nr = load_inode_bitmap (sb, block_group);
if (bitmap_nr < 0)
goto error_return;
-
+
bh = sb->u.ext2_sb.s_inode_bitmap[bitmap_nr];
-
- is_directory = S_ISDIR(inode->i_mode);
-
- /* Do this BEFORE marking the inode not in use */
- clear_inode (inode);
/* Ok, now we can actually update the inode bitmaps.. */
if (!ext2_clear_bit (bit, bh->b_data))
--
Andreas Dilger \ "If a man ate a pound of pasta and a pound of antipasto,
\ would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/ -- Dogbert
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu May 31 2001 - 21:00:41 EST