Messages in this thread | | | Date | Mon, 20 Mar 2000 23:30:23 -0500 | Subject | Re: Patch to make ext2 mounts go faster.... | From | tytso@mit ... |
| |
Date: Mon, 20 Mar 2000 16:57:25 +0100 (MEZ) From: "Dr. Michael Weller" <eowmob@exp-math.uni-essen.de>
> No, it's best to have extensive checks in user space rather than simple, > yet expensive, checks in the kernel. The checks have been there forever > and haven't caught much that isn't caught by the dirty fs flag.
Sorry to jump into this threat. But I have to disagree wholeheartedly!!!!
I experienced myself certain cases of buggy scsi drivers, broken hardware (scsi, motherboard or memory) wrong setup (dual boot machines with MS Win and linux and wrong partition setup) where exactly this silly consistency check came up and pointed out that something odd happened with the disk. The dirty flag would not deal with that issue.
However, for the majority of desktop linux installations with not that experienced users (I don't dare to call them admins, although they are the admins), these tests are sensible and helpful (and the bootup delay caused is not that high too) and need to be default because they'd not able them otherwise (because they would not know they should and how).
There are only two problems with your thesis:
(1) The check, although it is quite time consuming, only checks the bitmaps blocks to see if they match with the superblock group descriptors. On filesystems with 4k blocks, this means that you only catch errors that occur in 0.009% of the filesystem. (On filesystems with 1k blocks, you're cecking 0.03% of the filesystem --- in both cases, it's well less than a tenth of a percent.) Granted, the bitmap blocks get written a lot, but it still means that a large number of potential disk corruptions might not get noticed by the check.
(2) If there are problems, the kernel simply prints a warning. Many of the non-experienced users are likely to not notice or ignore the warning messages which appear at boot-time anyway.
(3) If someone *really* wants to do a potentially time-consuming minimalistic check, I can add it to e2fsck (it's basically means skipping passes 1 through 4 and only doing pass #5), but quite frankly, I'm still not convinced its worth the cost/benefit ratio. Still, people who are really paranoid could use this if they really want.
- 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/
| |