Re: How efficient is mounting /usr ro?
On Tue, 25 Nov 2003 21:14:21 +1100
Russell Coker <russell@coker.com.au> wrote:
RC> On Tue, 25 Nov 2003 19:51, Chema <chema.news.gmane@zarco.cjb.net>
RC> wrote:
RC> > Making /usr read-only is not for that kind of security. It will
RC> > keep your data safe from corruption (soft one, anyway: a disk
RC> > crash will take anything with it ;-). Besides, you can get a
RC> > better performance formating it with ext2, since you'll not need
RC> > journaling.
RC>
RC> Why would you get better performance? If you mount noatime then
RC> there's no writes to a file system that is accessed in a read-only
RC> fashion and there should not be any performance issue.
Hum, ¿are you talking only about ext3? 'Couse I don't think the reading performance of ext2 and reiserfs/jfs/whatever will be the same just by freezing the access time. Any test will tell you that they are not in usual conditions, e.g. http://fsbench.netnation.com/. ext3 is just a somewhat dirty hack on ext2, and without journaling their performance would be probably the same.
Now, how much difference really makes noatime??
Also, access time is usually a piece of information I'll like to keep. Probably some programs (maybee popularity-contest) would also like to know what is being touched.
Reply to: