You are not logged in.
After upgrading to lastest util-linux-ng (2.17.1-1) from 2.17.1 "fsck" always hangs at boot. I downgraded to 2.17.1 and it worked again. No error messages are output when it hangs. System is a Dell XPS M1530 laptop.
Thanks.
--mark
kernel26 2.6.32.9-1
Linux laptop-al 2.6.32-ARCH #1 SMP PREEMPT Tue Feb 23 19:43:46 CET 2010 x86_64 Intel(R) Core(TM)2 Duo CPU T7700 @ 2.40GHz GenuineIntel GNU/Linux
Last edited by markz79 (2010-03-08 13:09:25)
Offline
We have a big tracker: http://bugs.archlinux.org/
Offline
Seems to be working here, forced a fsck with 'sudo touch /forcefsck' and it checked all partitions (ext4) without a problem.
R00KIE
Tm90aGluZyB0byBzZWUgaGVyZSwgbW92ZSBhbG9uZy4K
Offline
I should metion that I'm using ext3, although not sure if it matters. Seems fsck hangs before it can invoke fsck.<fstype>.
Offline
I guess related to: http://bugs.archlinux.org/task/18579 Please see the explanation.
Offline
I do have rootfstype=ext4 in the kernel line so I guess that may help, I have never had any problems like the one you mention.
R00KIE
Tm90aGluZyB0byBzZWUgaGVyZSwgbW92ZSBhbG9uZy4K
Offline
Checked out the bug djgera referenced. Wasn't able to resolve by reverting to prev. version then upgrading again. I will double-check that rootfstype option doesn't resolve issue.
Offline
Still hangs with rootfstype=ext3 kernel option. Changed FSCK_ERR=/dev/stdout to make sure no errors are emitted, and none are. Strange.
Offline
Try to pass manually fsck and see how respond. (I assume that you don't use any encryption, lvm setup, just the standard)
boot with init=/bin/bash
/bin/mount -n -t tmpfs udev /dev -o mode=0755
/bin/mount -n -t proc none /proc
/bin/mount -n -t sysfs none /sys
/bin/cp -a /lib/udev/devices/* /dev/
echo > /proc/sys/kernel/hotplug
/sbin/udevd --daemon
/sbin/udevadm trigger
/sbin/udevadm settleskip all above commands if you have devtmpfs (not by default in Arch Linux kernel26) only need this:
mount -n -t devtmpfs none /devNow run a fsck -fv on all your partitions.
EDIT: use openvt is you need more consoles, run before fsck ![]()
Last edited by djgera (2010-03-08 00:17:58)
Offline
I saw blkid hang for me. I wonder if that's what's making fsck hang - that it calls blkid and waits forever on it. To solve that, I edited /etc/blkid.tab and deleted lines for anything that didn't need to be there - in my case, I removed entries for loop devices and that fixed it. Probably just removing that whole cache file would have been good enough.
I haven't tried to reboot yet since this update, so I don't know if I'm affected by the fsck problem itself.
Offline
Djgera: Thanks, tried what you suggested. Could do the last mount though as fstype "devtmpfs" wasn't recognized. Was still able to do a fsck and it hung.
Ataraxia: I do have some stale entries in blkid.tab. I'll remove them and see what happens.
Offline
@markz79: devtmpfs as said before, only works with a custom kernel with this option enabled it, Arch Linux kernel26 does not include it. You are welcome ![]()
About blkid.tab, interesting, is safe to delete it, is just a cache that blkid generate it for each device that "touches". Wating if this was the problem.
Last edited by djgera (2010-03-08 01:17:22)
Offline
Ataraxia:
That was it! Had stale references to a couple NTFS recovery partitions. "fsck" is doing fine now. Thanks!
Offline
Great! ![]()
Offline
@Djgera: That makes sense -- misread it at first. Thanks for your help.
Apparently blkid.tab can be deleted. I moved it and there were no problems.
Offline
No problem, please rename add [SOLVED] to title ![]()
Offline