Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Stuck in FSCK on boot ; Unable to allocate
16 views
Skip to first unread message
Beracah
Sep 24, 2012, 1:57:35 PM9/24/12
to
Hi all,
I've done a number of searches on this topic but failed to find anything relevant. Basically on boot on my headless server (1,000 miles away), a forced fsck is getting stuck in an infinite loop:
/dev/md3: Unable to allocate (# of blocks between 1 and 512) contiguous blocks for inode group (# range from 5600 to 5800)
I assume that this is because the disk is mostly full for what fsck wants to do. Before reboot, EXT4 was reporting i/o errors and inode count mismatches in dmesg. That said, both the raid group reported clean, smartctl no errors, and badblocks came back clean on both drives in the raid as well.
Does anyone have any suggestions for how to get this machine up and to clear some space, given that if it does come up the device will be mounted read-only?
Tech support reports that no interactive prompt appears, that fsck just gets to 8% completion, then cycles those errors and starts again at 0%.
Thanks,
Beracah
I've done a number of searches on this topic but failed to find anything relevant. Basically on boot on my headless server (1,000 miles away), a forced fsck is getting stuck in an infinite loop:
/dev/md3: Unable to allocate (# of blocks between 1 and 512) contiguous blocks for inode group (# range from 5600 to 5800)
I assume that this is because the disk is mostly full for what fsck wants to do. Before reboot, EXT4 was reporting i/o errors and inode count mismatches in dmesg. That said, both the raid group reported clean, smartctl no errors, and badblocks came back clean on both drives in the raid as well.
Does anyone have any suggestions for how to get this machine up and to clear some space, given that if it does come up the device will be mounted read-only?
Tech support reports that no interactive prompt appears, that fsck just gets to 8% completion, then cycles those errors and starts again at 0%.
Thanks,
Beracah
J.O. Aho
Sep 24, 2012, 3:30:39 PM9/24/12
to
system, this requires you to boot up from a rescue media, take a look at the
following article:
http://computer-forensics.sans.org/blog/2011/06/14/digital-forensics-mounting-dirty-ext4-filesystems
As things are as they are, there is always a risk you will loose some date or
corrupt things even worse, if you can use dd to make a copy of the media to
another device before you do this.
--
//Aho
Jasen Betts
Sep 25, 2012, 4:51:34 AM9/25/12
to
On 2012-09-24, Beracah <ber...@mit.edu> wrote:
> Hi all,
> I've done a number of searches on this topic but failed to find anything relevant. Basically on boot on my headless server (1,000 miles away), a forced fsck is getting stuck in an infinite loop:
>
> /dev/md3: Unable to allocate (# of blocks between 1 and 512) contiguous blocks for inode group (# range from 5600 to 5800)
>
> I assume that this is because the disk is mostly full for what fsck wants to do. Before reboot, EXT4 was reporting i/o errors and inode count mismatches in dmesg. That said, both the raid group reported clean, smartctl no errors, and badblocks came back clean on both drives in the raid as well.
>
> Does anyone have any suggestions for how to get this machine up and to clear some space, given that if it does come up the device will be mounted read-only?
plug an external drive in, (peferably eSATA) boot a "live disk", copy
> Hi all,
> I've done a number of searches on this topic but failed to find anything relevant. Basically on boot on my headless server (1,000 miles away), a forced fsck is getting stuck in an infinite loop:
>
> /dev/md3: Unable to allocate (# of blocks between 1 and 512) contiguous blocks for inode group (# range from 5600 to 5800)
>
> I assume that this is because the disk is mostly full for what fsck wants to do. Before reboot, EXT4 was reporting i/o errors and inode count mismatches in dmesg. That said, both the raid group reported clean, smartctl no errors, and badblocks came back clean on both drives in the raid as well.
>
> Does anyone have any suggestions for how to get this machine up and to clear some space, given that if it does come up the device will be mounted read-only?
everything to the external drive, reformat, copy everything you want back,
reinstall the bootloader. reboot.
--
⚂⚃ 100% natural
Jasen Betts
Sep 25, 2012, 7:30:52 AM9/25/12
to
debugfs I think it's called perhaps you can use it to nuke
/var/log/messages.2.gz or some other large unimportant file.
my impression is that you pretty much need to be a filesystem ninja
to know how to use it, and applying it against your only copy of the
data could be risky. If you had usable backup you'd be running it now
so this is probably not the solution.
--
⚂⚃ 100% natural
--- news://freenews.netfront.net/ - complaints: ne...@netfront.net ---
0 new messages