Every time I boot, I get primary superblock features different from backup, check forced on my root filesystem. There is an on-going thread on linux.debian.bugs.dist that is relevant. http://groups.google.com/group/linux.debian.bugs.dist/browse_thread/thread/3aafcdfe484d9852/409e330ad322db97 Here is the diff of "dumpe2fs -h /dev/hda1 >/tmp/hda1.super" and "dumpe2fs -ob8193 -oB1024 -h /dev/hda1 >/tmp/hda1.backup": diff -u /tmp/hda1.super /tmp/hda1.backup --- /tmp/hda1.super 2007-12-14 11:27:35.105284705 -0800 +++ /tmp/hda1.backup 2007-12-14 11:35:56.876966681 -0800 @@ -3,17 +3,17 @@ Filesystem UUID: 820bf306-09fd-45d7-b708-bbc209f3a5c7 Filesystem magic number: 0xEF53 Filesystem revision #: 1 (dynamic) -Filesystem features: has_journal ext_attr dir_index filetype needs_recovery sparse_super +Filesystem features: has_journal ext_attr dir_index filetype sparse_super Filesystem flags: signed directory hash Default mount options: (none) -Filesystem state: clean +Filesystem state: not clean Errors behavior: Continue Filesystem OS type: Linux Inode count: 7936 Block count: 31720 Reserved block count: 1586 -Free blocks: 22042 -Free inodes: 7882 +Free blocks: 11827 +Free inodes: 7871 First block: 1 Block size: 1024 Fragment size: 1024 @@ -22,9 +22,9 @@ Inodes per group: 1984 Inode blocks per group: 248 Filesystem created: Wed Nov 2 01:49:36 2005 -Last mount time: Fri Dec 14 11:12:20 2007 -Last write time: Fri Dec 14 11:12:20 2007 -Mount count: 4 +Last mount time: Mon Dec 10 04:39:08 2007 +Last write time: Mon Dec 10 04:39:08 2007 +Mount count: 0 Maximum mount count: -1 Last checked: Mon Dec 10 04:39:08 2007 Check interval: 0 (<none>) Note that the primary superblock has the attribute "needs_recovery" set. (This, after a recent reboot & fsck.) I am reverting to e2fsprogs-1.40.1 to see if the problem disappears and will report back.
Arrgh. [moron] /dev/hda1 is not my root filesystem. It's /boot. Here is the diff for /dev/hda3: diff -u /tmp/hda3.super /tmp/hda3.backup --- /tmp/hda3.super 2007-12-14 12:19:04.000873898 -0800 +++ /tmp/hda3.backup 2007-12-14 12:20:02.082373059 -0800 @@ -3,17 +3,17 @@ Filesystem UUID: 1f5eaa83-83d7-49c3-b21e-76bb225b2416 Filesystem magic number: 0xEF53 Filesystem revision #: 1 (dynamic) -Filesystem features: has_journal ext_attr dir_index filetype needs_recovery sparse_super +Filesystem features: has_journal ext_attr dir_index filetype sparse_super Filesystem flags: signed directory hash Default mount options: (none) -Filesystem state: clean +Filesystem state: not clean Errors behavior: Continue Filesystem OS type: Linux Inode count: 12091392 Block count: 24163398 Reserved block count: 1208169 -Free blocks: 2799067 -Free inodes: 10567707 +Free blocks: 2799699 +Free inodes: 10567726 First block: 0 Block size: 4096 Fragment size: 4096 @@ -22,9 +22,9 @@ Inodes per group: 16384 Inode blocks per group: 512 Filesystem created: Wed Nov 2 01:49:55 2005 -Last mount time: Fri Dec 14 11:56:49 2007 -Last write time: Fri Dec 14 11:56:49 2007 -Mount count: 2 +Last mount time: Mon Dec 10 06:30:13 2007 +Last write time: Fri Dec 14 10:53:14 2007 +Mount count: 0 Maximum mount count: -1 Last checked: Fri Dec 14 10:53:14 2007 Check interval: 0 (<none>) @@ -33,7 +33,6 @@ First inode: 11 Inode size: 128 Journal inode: 8 -First orphan inode: 9388661 Default directory hash: tea Directory Hash Seed: 118b6e72-dda7-491a-8e8c-79daf6d06e76 Journal backup: inode blocks I just rebooted, and I did not go through an fsck, but because of my hda1/hda3 confusion, I don't know if the "needs_recovery" attribute was set before my last reboot. I will reboot again and check.
dont bother posting information here. Theodore Tso is tracking the debian list and since he's the upstream e2fsprogs maintainer, we'll just grab the fix he posts.
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=454926#55 Looks like he has posted the fix.
which is already part of a release in the tree
I'm probably missing something, but looking at the changelog I can't see what revision you're talking about. Also, I'm experiencing this bug with the x86 stable 1.40.4. So, could you please be a little more specific ?