Date: Thu, 10 Sep 2015 08:57:44 -0400 From: Paul Kraus <paul@kraus-haus.org> To: FreeBSD Questions <freebsd-questions@freebsd.org> Subject: Re: Storage question Message-ID: <4E5C9688-700F-460D-907C-0EE2DB65CE08@kraus-haus.org> In-Reply-To: <55F10715.3040508@sneakertech.com> References: <55EF3D23.5060009@hiwaay.net> <20150908220639.20412cbd@gumby.homeunix.com> <55EF5409.8020007@yahoo.com> <55EFC2DA.3020101@hiwaay.net> <5EB5C2C2-575B-40BD-BF6A-85F396C058FE@kraus-haus.org> <55F058FC.6080204@hiwaay.net> <B737A47C-FECD-4882-9B57-DDF1B29FCDDF@kraus-haus.org> <55F09924.2000007@hiwaay.net> <67BE98EB-AD0F-46C2-AF26-02C65D828DCF@kraus-haus.org> <55F105B2.2080806@sneakertech.com> <55F10715.3040508@sneakertech.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sep 10, 2015, at 0:29, Quartz <quartz@sneakertech.com> wrote: >>> Note also the “dnr” datasets :-) >> >> Btw, you seem to be going with the 'dummy dataset' method for preventing >> fs fillup. What's the advantage of doing it this way vs refreservation >> or any other method? > > More specifically, dummy datasets strike me as kind of the ugly way of handling this issue. Isn't it better to just set parameters on your real datasets? I do it this way for two reasons. 1) Habit… I have been doing this since before zfs even had the refreservation and refquota properties :-) I have been managing zfs in production since zpool version 10 (clearly on Solaris since it has not been ported anywhere else that early on). 2) Transparency… it is much easier to spot a “dnr” dataset set than realize that a refreservation is set on the base zpool/dataset. Note that if you use refreservation, you only need it on the top level dataset in the zpool. -- Paul Kraus paul@kraus-haus.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4E5C9688-700F-460D-907C-0EE2DB65CE08>