Wrong filesystem line in /etc/fstab, no error message on bootscreen

[tabasko]

Binary package hint: mountall

First I wrote line to /etc/fstab, because I planned format my non-os parition filesystem from hfs+ to ntfs:
/dev/sda3 /media/Varasto vfat rw,exec,auto,users 0 0

I bootted to OSX (I have OSX and Ubuntu lucid dualboot) and did change my mind and format partition to fat32, not ntfs. When I did try to boot my Ubuntu after this, it did hang to plymouth bootlogo for long time, there was no error message before I hit Esc-key. Then there was scrolling errors from mountall with cryptic numbers.
I fixed my /etc/fstab line to
/dev/sda3 /media/Varasto vfat rw,exec,auto,users 0 0

And Ubuntu did boot well.

It would be good, if mountalls error would show rightaway on bootscreen.

Description: Ubuntu lucid (development branch)
Release: 10.04
mountall:
  Asennettu: 2.11
  Ehdokas: 2.11
  Versiotaulukko:
 *** 2.11 0
        500 http://fi.archive.ubuntu.com/ubuntu/ lucid/main Packages
        100 /var/lib/dpkg/

[Scott James Remnant (Canonical) (canonical-scott)]

Could you please add those "cryptic numbers" to the bug report? They could be important!

Note that it's intended that an error mounting a filesystem *is* output to the boot screen, with a choice of how to proceed. Since this isn't happening for you, I will need more details to debug your problem.

Thanks

[Marcos Agüero (wiredrat)]

I've got the same problem with incorrect lines in fstab (bad device or mount point name). ie:

/dev/nonexistan /home ext3 defaults,auto 0 0

It is annoying on a server system because it never boots (and this happen to me during a Landscape schelude reboot!!!). Would be nice have a timeout option for non-essentials mounts.
In my case, when I plugged a monitor, didn't show a error message neither a prompt (may be due a wrong monitor resolution?). So I needed to debug whole boot process to find out that it was a mountall problem.
I could boot with init=/bin/sh kernel param and fix fstab manually.

[Dave Gilbert (ubuntu-treblig)]

Confirming because I'm seeing this on current Maverick (as of today).

I added a line to the end of my fstab:

/dev/sde /frob default ext3 defaults 1 2

(sde doesn't exist).

Mountall just hangs - no diagnostics at all. Rebooting in 'recovery mode' doesn't help either.
A missing drive or screwed up /etc/fstab should be fixable from recovery mode shouldn't it?

In the recovery mode my console says:

fsck from util-linux-ng 2.17.2
fsck from util-linux-ng 2.17.2
/dev/vda1 was not cleanly unmounted, check forced.
/dev/mapper/server1-root: clean, 81728/874832 files, 460278/3495936 blocks (check in 5 mounts)
/dev/vda1: 225/124496 files (7.1% non-contiguous), 73373/248832 blocks
mountall: fsck /boot [346] terminated with status 1

and that's it

So the only diags were from the unclean reboot (not sure why it's moaning about /boot) and no diags at all for the broken/wrong device.

Dave

[Dave Gilbert (ubuntu-treblig)]

Hmm, OK, mine is the same as bug 593916 - so I guess Tabasco you do need to answer what the cryptic numbers were - were they just the same type of debug I list or were they something else?

Dave

[GerhardGaußling (ggrubbish-web)]

I also find it very annoying, that mountall doesn't give helpfull diagnoastic messages at all if a boot device is not available or a fstab line is faulty. It not even droppes someone to a shell, not to mention, that it doesn't give the hint to press CTRL+ALT+SysRq+K to get a maintenance shell. That's insane. It's very annoying too, that it doesn't show a progressbar on fsck, especially on big harddrives.

[Luigi Molinaro (luigi-molinaro)]

Just type 's' to skip, and login and modify fstab :)

[Launchpad Janitor (janitor)]

[Expired for mountall (Ubuntu) because there has been no activity for 60 days.]