Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.mira.net.au!news.mel.connect.com.au!munnari.OZ.AU!spool.mu.edu!howland.reston.ans.net!cs.utexas.edu!uwm.edu!math.ohio-state.edu!jussieu.fr!eurocontrol.fr!ramses.eurocontrol.fr!not-for-mail From: rbt@eurocontrol.fr (Ollivier Robert) Newsgroups: comp.unix.bsd.freebsd.misc Subject: Re: Increasing Inodes Date: 11 Jul 1996 09:03:32 GMT Organization: Eurocontrol EEC, Bretigny, France Lines: 21 Message-ID: <4s2g14$je0@ramses.eurocontrol.fr> References: <4s29ij$ri7@liberator.concentric.net> NNTP-Posting-Host: mozart.eurocontrol.fr [courtesy cc of this posting sent to cited author via email] In article <4s29ij$ri7@liberator.concentric.net>, Damien Thorn <Judasg@cris.com> wrote: > Also, the drives have (and are probed as) having 512 byte sectors. > FreeBSD is defaulting to using 1k blocks. Should we redo the FS with > 512 byte blocks as well? Does this require even more inodes? You're talking about two things here. The physical sector size is 512 bytes but it has nothing really to do with the "fragment" size of 1 KB you've seen. It seems that you've taken the default size of 8 KB/1 KB (full block is 8 KB and fragment size is 1 KB). For Usenet disks, you'd better use 4 KB/512 bytes. As for the # of inodes per FS, it is 4x the fragment size. So it will be 1 inode per 2 KB for 4 KB/512 FS and 1 inode per 4 KB for 8 KB/1 KB FS. The man page for newfs(8) is not clear and it has been corrected in CURRENT. -- Ollivier ROBERT -=- Eurocontrol EEC/TIS -=- Ollivier.Robert@eurocontrol.fr Usenet Canal Historique