Re: XFS feature request

[Seth Mos <knuffie@xxxxxxxxx>]

At 17:29 11-6-2003 -0600, Michael Loftis wrote:
> True...There may be situations where having the filesystem degrade into 
> nothing more than a semi-useful source of random entropy for PRNG may be ok :)
>
> --On Wednesday, June 11, 2003 4:32 PM -0500 Austin Gonyou 
> <austin@xxxxxxxxxxxxxxx> wrote:
>
> > Given the path of comments here. I'd say that's why the ext2 FS has 3
> > options for error handling! :)

The rootfs should remount ro so the box doesn't die instantly and takes 
down the entire machine.

This is exactly what happened here once. The duplicate inode was not found 
by tar but xfsdump did a bulkstat and noticed it. Because of this the 
rootfs unmounted and caused all 40 users to be effectively kicked off the 
machine which is not very nice.

In my opinion the rootfs should be the last to dissapear and be remounted 
ro instead.

If the other volumes take a nose dive that's ok, but I _need_ to have at 
least my root filesystem to fix something.

In this case the corruption must have been there for at least 6 months or 
so and we didn't even notice. The rootfs is completely backuped every day 
by tar which was not enough for XFS to notice it was there.

Cheers

--
Seth
It might just be your lucky day, if you only knew.