format/newfs larger external consumer drives

Andrew Duane aduane at juniper.net
Fri Jul 24 12:49:44 UTC 2015


> -----Original Message-----
> From: owner-freebsd-hackers at freebsd.org [mailto:owner-freebsd-
> hackers at freebsd.org] On Behalf Of Don whY
> Sent: Thursday, July 23, 2015 8:52 PM
> To: freebsd-hackers at freebsd.org
> Subject: Re: format/newfs larger external consumer drives
> 
> On 7/22/2015 2:08 PM, Dieter BSD wrote:
> > Don whY asks:
> >> So, fsck's effort (and execution *time*) is based *mostly* on inodes?
> >
> > I don't know about *mostly*, but reducing the number of inodes
> > significantly reduced fsck time for me.
> 
> OK.  I may try building a filesystem, loading a fixed set of files
> (assorted) onto it, then fsck'ing it.  Then, rebuild with a different
> block/frag/inode configuration and try again (same file set).  At the very least, it
> will be an interesting experiment!

FSCK has 5 passes, each of which checks a different "thing" in the filesystem. Each pass will depend on how many of those "things" there are to check. One checks inodes and cylinder groups, one checks all blocks, one checks all directory and file entries, and so on. So fewer inodes will help, but more files eat up the savings, etc. Some experiments can tell you how long it takes to check each and what the savings would be.

Also remember that each inode takes 128 bytes on disk. So a billion unused inodes wastes a 128 GB on the disk.

> _______________________________________________
> freebsd-hackers at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
> To unsubscribe, send any mail to "freebsd-hackers-unsubscribe at freebsd.org"


More information about the freebsd-hackers mailing list