0

I have just installed my Ubuntu 16.04 few weeks earlier and it freezes each time while suspend and hibernate, I have to force power off the system. On boot dev\sda1: recovering journal clearing orphaned inode "some inode numbers" and then it boots up. Here is some info:

prasha@prasha:~$ sudo blkid

/dev/sda1: UUID="b21ebca2-7efd-4bb5-95fb-a06863eb9f01" TYPE="ext4" PARTUUID="616adc3d-01"
/dev/sda5: UUID="683ffa03-a7fc-483c-b4e3-5b2c95c54c24" TYPE="swap" PARTUUID="616adc3d-05"

prasha@prasha:~$ cat /etc/initramfs-tools/conf.d/resume

RESUME=UUID=683ffa03-a7fc-483c-b4e3-5b2c95c54c24
0

The first thing is to perform a file system check...

  • boot to the GRUB menu (where you select which OS to start)
  • choose Advanced Options
  • choose Recovery mode
  • choose Root access
  • at the # prompt, type sudo fsck -f /
  • you can ignore errors about the time/date/superblock
  • repeat the fsck command if there were other errors
  • type reboot

The second thing to do is to use the SMART internal disk monitoring tools to see if it detects a pending hard disk failure...

  • start the Disks application from the Unity dash
  • select the hard disk in the left pane
  • click on the "hamburger" icon in the top-right of the Disks window
  • select SMART Data & Tests
  • view the data, and run the tests
0

I recommend running some hardware tests. Start with memfree86 to test your memory. Then move on to the smartmontools package to read the S.M.A.R.T. data of your hard disk.

Run memfree86

  1. Restart you computer.
  2. During startup push ESC key to show the Grub boot menu.
  3. Choose memfree86 memory entry and give it time to test a few passes of your memory.

NOTE: This step can take several minutes to an hour per pass depending on amount of RAM. The program will test over and over again until you cancel it. The UI notes the number of passes; that is the number of times it has tested all of your RAM.

Run smartmontools

Assuming your RAM is okay, boot to an Ubuntu Live CD. Install the package.

apt-get install smartmontools

Typically hard drives are located at /dev/sda. Run the smartctl command to see the health of your drive.

smartctl -a /dev/sda

The wikipedia page on S.M.A.R.T. does a good job of pointing out values you need to pay attention. You should look for errors typically in the magnitude of 1k-100k error events. On my drives, I've seen drives have as low as zero for all error events over the course of 5+ years usage.

Next steps

If your RAM shows errors then pull out your RAM sticks and test them one at a time to determine the bad SODIMM. Replace any bad RAM stick.

If your hard drive is showing a lot of concerning errors then I recommend immediately recovering you data to a known healthy disk from a Live CD. The unhealthy disk should be replaced with a healthy one.

If both the RAM and disks show up as completely healthy then move on to checking software (such as fsck).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.