Messages in this thread | | | Date | Thu, 12 Nov 1998 22:49:55 GMT | From | "Theodore Y. Ts'o" <> | Subject | Re: e2fsck... |
| |
Date: Thu, 12 Nov 1998 16:52:39 +0100 (MET) From: Riccardo Facchetti <fizban@tin.it>
On Thu, 12 Nov 1998, Mofeed Shahin wrote:
> So let me get this straight, If I want to check my file system for errors > there is no problems, but if I want to fix any errors found I have to > reboot via a boot disk, and then e2fsck the now unmounted drive ?
Moreover, checking a filesystem for errors (using e2fsck -n) while the filesystem is mounted may falsely show that there are problems when the filesystem is in reality OK. That's because the kernel may be in the middle of modifying the filesystem while e2fsck (which is a multi-pass operation) is running, and so it will find a discrepancy which is simply caused by the fact that the kernel modifying the filesystem while e2fsck isn't looking.
And, if you let e2fsck actually modify the filesystem that's mounted, and it tries to correct the false "inconsistency", the result may be a badly corrupted filesystem.
Multics did allow on-line salvaging of volumes, and it is theoretically possible to do such a thing. But it requires that the filesystem checker lock out kernel changes, and the filesystem checker would have to be in close communication with the kernel while it went about its business. E2fsck and the kernel both don't have the necessary infrastructure to do this today.
- Ted
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |