Date: Tue, 26 Jun 2012 19:29:42 +0300 From: Andriy Gapon <avg@FreeBSD.org> To: Mark Felder <feld@feld.me> Cc: freebsd-fs@FreeBSD.org Subject: Re: [PATCH] Simple ARC stats in top Message-ID: <4FE9E376.8030209@FreeBSD.org> In-Reply-To: <4FE9DE92.3040700@FreeBSD.org> References: <201206251443.41768.jhb@freebsd.org> <op.wgik62ia34t2sn@tech304> <4FE9CC00.9090501@FreeBSD.org> <op.wgins2ez34t2sn@tech304> <4FE9DB60.1030905@FreeBSD.org> <op.wgio9gg934t2sn@tech304> <4FE9DE92.3040700@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
on 26/06/2012 19:08 Andriy Gapon said the following: > So the correct ARC size breakdown is: hdr_size + data_size + other_size + > l2_hdr_size, where data_size can further be broken down to mfu_size + mru_size + > anon_size. > > I think that users of zfs-stats / arc_summary.pl could benefit from this knowledge > too. > (I will repeat just in case that zfs-stats mistreats kstat.zfs.misc.arcstats.p as > current MRU size whereas it is actually target MRU size) Here's a quick attempt at reflecting my understanding of the stats in zfs-stats code: http://people.freebsd.org/~avg/zfs-stats I think that this variant of the script should produce results more aligned with what top shows with John's patch. -- Andriy Gapon
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4FE9E376.8030209>