Date: Mon, 11 Jun 2007 23:40:22 +0200 From: Ivan Voras <ivoras@fer.hr> To: freebsd-fs@freebsd.org Subject: Re: Inode density for database machines - newfs -i Message-ID: <f4kfg8$3t2$1@sea.gmane.org> In-Reply-To: <cone.1181595086.390147.8020.5001@35st.simplicato.com> References: <cone.1181531481.451392.13436.1000@zoraida.natserv.net> <20070611161148.GA19299@keira.kiwi-computer.com> <cone.1181585492.585345.8020.5001@35st.simplicato.com> <f4kadf$g00$1@sea.gmane.org> <cone.1181595086.390147.8020.5001@35st.simplicato.com>
next in thread | previous in thread | raw e-mail | index | archive | help
[-- Attachment #1 --] Francisco Reyes wrote: > Ivan Voras writes: > >> How long fsck takes depends on the number of used inodes, not the total >> number. Barring a bug in UFS, you should be able to have as little >> inodes as you want, especially if you know for sure what number to >> expect. > > "man tuning" says otherwise. As Rick already clarified, this is a side-effect of UFS2. Actually, fsck for UFS2 will need to check all "initialized" inodes, e.g. if you had a file, then deleted it, that inode will still need to be checked, but never-initialized nodes won't. tuning(7) probably wasn't updated. [-- Attachment #2 --] -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.4 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGbcFGldnAQVacBcgRAhU2AJ9tYpUkcP0gEkcDOOEv5zM2T29oeACg8kiX 1+EFOIWAaw1Fj47TQz5KJ2U= =gOxX -----END PGP SIGNATURE-----
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?f4kfg8$3t2$1>