best practices using LVM and e2fsck

[Jarod Wilson]

On Thu, Jul 1, 2010 at 9:47 AM, Edward Ned Harvey <blu-Z8efaSeK1ezqlBn2x/YWAg at public.gmane.org> wrote:
>> From: discuss-bounces-mNDKBlG2WHs at public.gmane.org [mailto:discuss-bounces-mNDKBlG2WHs at public.gmane.org] On
>> Behalf Of Tom Metro
>>
>> I don't recall the ext version being mentioned in this thread. Your
>> concern seems to imply v2. Would switching to v3 or v4 be an option,
>> which should eliminate the possibility of a long fsck run?
>
> In ext2, you must fsck every time there's an ungraceful dismount.
> In ext3/4, you can avoid fsck's after ungraceful dismounts, but still, once
> in every ... something like 180 days or 90 reboots or something like that
> ... It will still fsck during startup.

...if you leave outdated e2fsprogs default format options in place. :)

Seriously, newer e2fsprogs than what was originally in RHEL5-based
distros actually disables the forced fsck on journaled file systems,
if I'm thinking clearly. Or perhaps its an installer-level tweak for
newer Fedora/RHEL, can't recall. But I'm fairly certain it was
determined a year or three ago that forced fsck was not really
something one should have by default any longer.

-- 
Jarod Wilson
jarod-ajLrJawYSntWk0Htik3J/w at public.gmane.org