Check a mounted disk and fix errors on Amazon EC2

When I logged into my EC2 instance, I had the warning:

*** /dev/xvda1 should be checked for errors ***

We can't fsck /dev/xvda1 because it is mounted and in use, so I need to check this disk on startup as it is mounted, however there are a couple of things that need to be done first before you can do this on my Ubuntu EC2 instance.

First I need to enable FSCKFIX, which is off by default

sudo nano /etc/default/rcS

find the line:
#FSCKFIX=no

And change it to:

FSCKFIX=yes

Then I need to

sudo nano /etc/fstab

In my instance there was only one record, the one for "/" and its last digit, the fsckorder was '0', which tells fsck to skip the disk, so I changed it change it to '1'.

I looked for the line:
LABEL=cloudimg-rootfs   /        ext4   defaults,discard        0 0

and changed it to:

LABEL=cloudimg-rootfs   /        ext4   defaults,discard        0 1

I then needed to create: the /forcefsck file so that fsck was run on next reboot:

sudo touch /forcefsck

I then rebooted the Server from ec2 console.

After the restart I logged on again I checked that the message was gone and reset all these values back to their originals

Add new comment