On Wed, Mar 24, 2010 at 05:29:58PM -0700, Elliott Mitchell wrote:
> Package: e2fsprogs
> Version: 1.41.3-1
>
> Subject tells the story. Why are mere /precautionary/ filesystem checks
> allowed to slow a system booting so much? ie the ones triggered by either
> the mount count or check time exceeding their limits.
>
> I would suggest limiting it to doing a precautionary check on *one*
> filesystem (or perhaps one filesysterm per drive) when booting. Also if a
> filesystem gets checked due to other issues, that should inhibit the
> precautionary check.
>
> This is particularly excessive for filesystems that only get mounted
> read-only. Read-only filesystems additionally do not need to halt the
> boot process when getting precautionary checks, since they can be checked
> in the background even while the kernel has them mounted.
If you are mounting the file system read-only, feel free to change the
fsck pass number in /etc/fstab to be 0. That will cause e2fsck to be
skipped completely.
Note that mount count is randomized these days, to try to avoid
multiple file systems from being checked in a single boot.
Also feel free to disable the precautionary checks if you don't want
to use it using tune2fs(8). If you are confident that your hardware
is perfect, and hard drive will never corrupt your metadata blocks,
the precautionary checks can easily be dsiabled.
- Ted
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
