fsck: Error determining size of the physical device: File too large

So, I got done patching this fairly important server, running RedHat 5.10. For reasons I will not go into, the server has about 30TB of direct attached storage. After the reboot, one of the filesystems, in this case it had 16TB of storage, failed to pass automatic fsck check on boot.

So, I booted it into single user mode to take a look:

[root@prdnb0001 dev]# fsck -n /dev/vgmd1200/optlv
fsck 1.39 (29-May-2006)
e2fsck 1.39 (29-May-2006)
Error determining size of the physical device: File too large

That did not look promising. I was little dumbfounded. I googled a few things, saw some not so helpful pages, but nothing 100% definite.

Nevertheless, I got a little bit of inspiration. By the way, my version of e4fsprogs was 1.41.12-3.el5. The filesystem was ext4 filesystem so I gave this a shot:

[root@prdnb0001 dev]# fsck.ext4 -n /dev/vgmd1200/optlv

…and filesystem check proceeded to run just fine. I did not have time to dig into why this worked and automatic fsck check on boot failed. Maybe someone can shed some light on this.

In any case, reading man pages always pays and once in a while you find out something cool such as:

SIGUSR1 This signal causes e2fsck to start displaying a completion bar or emitting progress information.

So by doing:

[root@prdnb0001 ~]# pkill -SIGUSR1 fsck.ext4

in another terminal window, I was able to get progress bar for the running check.

Posted on September 30, 2013 at 09:42 by somedude · Permalink
In: centos, ext4, linux, linux tips, redhat, storage

Leave a Reply