> > He said that ext2 fsck right *now* would take 3
> > hours for a 100GB filesystem. I replied saying that I'd benchmarked
> > it at 13 minutes for a 75% full 43GB filesystem (and I now have 4 such
> > filesystems attached to our mail server cluster). Within a year, I
> > expect to see journalling support for ext2. Within that year, I don't
> > expect any system crashes and, if there is a crash, I find 13 minutes
> > to be an acceptable fsck time given the other advantages of the Linux
> > mail cluster.
> I find the 5 minutes for my 75% full 6.4GB filesystem tedious but ok.
> If fsck time were linear, the same system would take 33.5 minutes for 43GB.
> Does this mean fsck times don't scale linearly with disk size?
Well - fsck time depends lineralily on _ammount of data in files
>13K_, for 1K filesystems. Fsck time is cut down quite a bit when you
use 4K blocksize.
So his report is probably right after all: with 4K blocks fsck tends
to be quite fast. (This is on 100Gig filesystem with 4K blocks and
mostly small files:)
root@D46:/work-save > time fsck.ext2 -n -tt /dev/sda3
e2fsck 1.14, 9-Jan-1999 for EXT2 FS 0.5b, 95/08/09
Warning! /dev/sda3 is mounted.
/dev/sda3 was not cleanly unmounted, check forced.
Pass 1: Checking inodes, blocks, and sizes
Pass 1: Memory used: 106496, elapsed time: 709.284/19.470/ 8.750
Pass 2: Checking directory structure
Pass 2: Memory used: 0, elapsed time: 49.570/ 6.010/ 2.670
Pass 3: Checking directory connectivity
Peak memory: Memory used: 176128, elapsed time: 758.915/25.480/11.440
Pass 3: Memory used: 0, elapsed time: 127.633/127.570/ 0.000
Pass 4: Checking reference counts
Pass 4: Memory used: 0, elapsed time: 0.784/ 0.780/ 0.000
Pass 5: Checking group summary information
Pass 5: Memory used: 0, elapsed time: 20.616/ 2.680/ 0.230
/dev/sda3: 1612171/3325952 files (0.1% non-contiguous),
Memory used: 176128, elapsed time: 907.942/156.510/11.670
So it is 1K blocksize which is a problem...
-- The best software in life is free (not shareware)! Pavel GCM d? s-: !g p?:+ au- a--@ w+ v- C++@ UL+++ L++ N++ E++ W--- M- Y- R+
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to firstname.lastname@example.org Please read the FAQ at http://www.tux.org/lkml/