1

Final Update:

I ended up replacing Ubuntu with Linux Mint and the problem of random crashes seems to have gone away (been using it heavy-duty for a good couple of hours by which time Ubuntu would have normally crashed on me).

I know that's not a great answer but I did try hard to figure out what was going on and came out with naught. Mint is also based on Ubuntu 16.04 which makes it even more puzzling.

Hopefully this is somewhat useful to anyone else that gets similar symptoms to what I was getting (where after using it for a while the hard-disk gets disconnected, the Ubuntu menu stops working, and a hard-reset is required).

Original post:

I keep getting errors when I try to run fsck. Ubuntu also keeps crashing on me.

I'm not convinced it's a hard drive issue as disk checks other than fsck do not report a problem and I've even shifted the Ubuntu partition around (see original post here: Dependency failed for /dev/disk/).

The error I get when running fsck in recovery mode is: Timed out waiting for device dev-disk-by\x2duuid-1d227bb3\x2db6ac\x2d44f6\x2dda063\ ...I could not read the rest

My fstab is:

# /etc/fstab: static file system information.
#
# Use 'blkid' to print the universally unique identifier for a
# device; this may be used with UUID= as a more robust way to name devices
# that works even if disks are added and removed. See fstab(5).
#
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
# / was on /dev/nvme0n1p7 during installation
UUID=06473276-9e10-42e4-86db-fc3c81557d8b /               ext4    errors=remount-ro 0       1
# /boot/efi was on /dev/nvme0n1p1 during installation
UUID=A48B-EA59  /boot/efi       vfat    umask=0077      0       1
# swap was on /dev/nvme0n1p6 during installation
UUID=1d227bb3-b6ac-44f6-a063-2bd5e8578eba none            swap    sw              0       0

My blkid is:

/dev/loop0: TYPE="squashfs"
/dev/loop1: TYPE="squashfs"
/dev/nvme0n1: PTUUID="06fdaa6e-9825-4df2-8d9b-b405e996ba04" PTTYPE="gpt"
/dev/nvme0n1p1: LABEL="SYSTEM" UUID="A48B-EA59" TYPE="vfat" PARTLABEL="Ubuntu Boot" PARTUUID="5402c9a1-2c4f-4dd4-9b56-e5068b5f0950"
/dev/nvme0n1p2: PARTLABEL="Microsoft reserved partition" PARTUUID="52ef0e33-bb36-4ee5-9b96-a8595743b369"
/dev/nvme0n1p3: LABEL="Windows" UUID="7ACA8CC9CA8C82DD" TYPE="ntfs" PARTLABEL="Basic data partition" PARTUUID="1b57fd79-f182-4786-b140-7f8b4688f5f6"
/dev/nvme0n1p4: UUID="0202928302927AFD" TYPE="ntfs" PARTUUID="9c06573d-1ecb-4821-abea-a51ab044f001"
/dev/nvme0n1p5: UUID="e681ae06-9f7e-4902-ae91-4d1f8dcf028d" TYPE="ext4" PARTLABEL="Basic data partition" PARTUUID="1a2ac5ca-37c5-11e7-9246-00155d6ff614"
/dev/nvme0n1p6: UUID="1d227bb3-b6ac-44f6-a063-2bd5e8578eba" TYPE="swap" PARTLABEL="Basic data partition" PARTUUID="1a2ac759-37c5-11e7-9246-00155d6ff614"
/dev/nvme0n1p7: UUID="06473276-9e10-42e4-86db-fc3c81557d8b" TYPE="ext4" PARTLABEL="Ubuntu" PARTUUID="1a2ac76e-37c5-11e7-9246-00155d6ff614"
/dev/nvme0n1p8: LABEL="Windows RE tools" UUID="5E348D1C348CF7F3" TYPE="ntfs" PARTLABEL="Basic data partition" PARTUUID="d7e07fd7-01ac-4e69-bf35-ad57fabdfd1c"
/dev/nvme0n1p9: LABEL="LinuxStorage" UUID="dc012449-3e39-4c52-97fb-1ffe9df2cab9" TYPE="ext4" PARTLABEL="LinuxStorage" PARTUUID="e9e0ec72-be01-436a-bbf7-f574a28f35fa"
/dev/sda1: LABEL="UBUNTU 16_0" UUID="F852-224C" TYPE="vfat" PARTUUID="00010c9d-01"

Does my fstab look ok to people?

UPDATE

Here is what I have found since:

  • if I go into root and wait a few seconds the system crashes/displays this on its own: https://www.dropbox.com/s/xzux7fj997dxobt/IMG_20170603_104418496.jpg?dl=0
  • For this error to appear I have to start typing out a command such as fsck or reboot or pretty much anything. It then crashes before I hit enter

  • Whatever command I had started typing out gets run after it finishes crashing, so reboot is the safest thing here

  • Because that error indicated it was my swap that was failing I formatted my swap and updated my fstab

  • However the above error returned. I was unable to get a good photo of the error as it only stays on the screen a few seconds before it disappears

  • But I can tell you the main error at the top seems to point to A48B now , which according to blkid is my Ubuntu Boot partition on https://www.dropbox.com/s/ztqw1yd8a7qy6t0/IMG_20170603_110604580.jpg?dl=0

  • Not sure what I should do with that as didn't want to mess around with boot. But I do have a USB boot I can use to format p1 perhaps?

Reddspark
  • 119
  • 1
    Matching the UUIDs, it looks OK. – muru Jun 01 '17 at 15:33
  • As muru says , your /etc/fstab file is OK. – Ali Razmdideh Jun 01 '17 at 16:13
  • 1
    It looks like you are trying to run fsck on swap 1d22.... UUID. The fsck file check is only for ext2, ext3 & ext4 formats. And swap is unformatted, so you cannot nor need to run fsck on it. I might add noatime as parameter for mounting in fstab for any ext4 partition. – oldfred Jun 01 '17 at 16:22
  • In terminal please show me the output of free -h and swapon and ls -alt /var/crash. Show me the exact fsck commands that you were using. Edit/paste the results into your question please. Start comments directed to me with @heynnema or I may miss them. – heynnema Jun 01 '17 at 18:08
  • Could you please upload your fstab file to a pastie or file sharing service? From the terminal output above I suspect that there are bogus non-printable characters embedded in the UUID that don't show up on a web site inside a Unicode-capable web browser. – David Foerster Jun 01 '17 at 22:02
  • Thanks appreciate the help ... About to head off to an exam so will respond tonight. I was just typing in fsck and letting it run on all drives in recovery mode. – Reddspark Jun 02 '17 at 10:00
  • Ok so firstly here is my latest fstab: https://www.dropbox.com/s/om6xip5ekzjjqwi/fstab?dl=0 – Reddspark Jun 03 '17 at 10:24
  • free -h gives: total used free shared buff/cache available Mem: 15G 1.4G 12G 463M 1.5G 13G Swap: 4.0G 0B 4.0G – Reddspark Jun 03 '17 at 10:24
  • swapon gives: NAME TYPE SIZE USED PRIO /dev/nvme0n1p6 partition 4G 0B -1 – Reddspark Jun 03 '17 at 10:24
  • ls -alt /var/crash gives:

    total 648 drwxrwsrwt 2 root whoopsie 4096 Jun 3 10:26 . -rw------- 1 whoopsie whoopsie 0 Jun 1 16:49 sbt.0.uploaded -rw-r--r-- 1 root whoopsie 0 Jun 1 16:49 sbt.0.upload -rw-r----- 1 root whoopsie 302627 Jun 1 16:49 sbt.0.crash -rw-r----- 1 root whoopsie 351062 Jun 1 15:31 _usr_lib_snapd-glib_snapd-login-service.0.crash drwxr-xr-x 14 root root 4096 Feb 15 20:43 ..

     – Reddspark Jun 03 '17 at 10:25
  • The exact fsck command I last ran was: fsck /dev/nvme0n1p5 -p -c
    The output I got from this (beore it the crashed) was 'updating bad block inode 11/100256 files'     – Reddspark Jun 03 '17 at 10:25
  • ...also PLEASE SEE UPDATE ABOVE – Reddspark Jun 03 '17 at 10:27

1 Answers1

0

Unless your machine to totally starved for DRAM, swap should be running there and not as a HDD/SDD partition. That said, you can check for HDD bad blocks with this command:

sudo badblocks -v /dev/sdax # where x = /dev/swap_partition_name_ID

Running fsck on the swap partition is not needed and is probably not even desirable.

As a practical matter, and to answer your concerns, a reasonably resourced machine will rarely if ever actually use swap.

Ali Razmdideh
  • 5,758
jones0610
  • 2,157
  • Don't use this on a SSD! – heynnema Jun 01 '17 at 18:04
  • I believe I said HDD not SSD – jones0610 Jun 01 '17 at 19:07
  • 1
    But nvme0n1p6 is on a SSD. And... the correct way to badblock anything is e2fsck -c -k /dev/sda from a Live DVD/USB. – heynnema Jun 01 '17 at 19:08
  • The OP stated that s/he was dealing with a hard disk issue. I presume s/he knows the difference between a HDD and SSD. The bottom line is that s/he should be running swap out of RAM. And if a HDD swap partition is suspect, the answer I gave will identify any problems. SSDs to my knowledge are impervious to bad blocks. I may have assumed a base knowledge of those reading this that may not be realistic in all cases. – jones0610 Jun 01 '17 at 19:15
  • 1
    The OP did reference a hard disk... but it's up to the person who answers the question to know that nvme0np6 is a SSD, and not to give a (incorrect... you don't run badblocks command directly... and read man badblocks for the warning about feeding the output to...) badblocks command (e2fsck is appropriate if HDD) against it. – heynnema Jun 01 '17 at 19:22
  • I appreciate the edits ali76. I could never make a living as a tech writer :) – jones0610 Jun 01 '17 at 19:52
  • Sorry yes it is a SSD. In my mind SDD was just a type of hard disk so was just calling it a hard disk. My bad. About to head off to an exam so will look through above comments tonight and respond. – Reddspark Jun 02 '17 at 09:57