[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Max Mount Count



Andrew writes:
> Offtopic for this list, but I note that when I set up an IDE
> partition as type 0x8e and run
> 
> vgscan
> pvcreate /dev/hde7
> vgcreate test /dev/hde7
> 
> I see "lvm -- lvm_chr_ioctl: unknown command 4004fe0a" in
> the logs. lvm_0.9.1_beta7.tar.gz, kernel 2.4.5.

That is an artifact of using LVM 0.9.1b2 kernel code (what is in the
Linus kernel) and 0.9.1b7 user tools, but is not a real problem.
Alan has been slowly merging the beta7 code into the -ac kernel, and
presumably will be feeding it to Linus at some time.

> ext3 seems to interoperate happily with LVM now.  I haven't
> tested striping and snapshots yet - I'm hoping Andreas will :)

I'm fundamentally against striping for 99% of cases.  Many people think
it will improve performance (i.e. "Kewl my disk will be twice as fast")
but in reality this is only true if you are streaming reads/writes at
the maximum bandwidth of the disks, and all of the upstream bandwidth
is twice as high...

I'd rather have two (or more) disks doing totally separate things
(e.g. /tmp, /usr/src/linux, /var) to avoid lots of seeks.  Also, striping
reduces the reliability of your system by half.

Cheers, Andreas
-- 
Andreas Dilger  \ "If a man ate a pound of pasta and a pound of antipasto,
                 \  would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/               -- Dogbert





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]