Date: Tue, 7 Dec 2010 17:18:11 -0800 From: David Brodbeck <gull@gull.us> To: freebsd-fs@freebsd.org Subject: Re: zfs snapshots. Message-ID: <AANLkTinrmKj6kJ1N0-Ro6%2BkA04_KKRXCvdECT45zVyqA@mail.gmail.com> In-Reply-To: <7554F69B-0E03-42E3-8BB2-FD7CFE19805A@pean.org> References: <4CF017C8-9D3C-483D-B508-5BF70E32AA14@pean.org> <4CF949B0.5060405@obsysa.net> <E629A6C6-2784-41B5-9979-A407464C52F0@pean.org> <20101203210045.GB7429@lor.one-eyed-alien.net> <7554F69B-0E03-42E3-8BB2-FD7CFE19805A@pean.org>
index | next in thread | previous in thread | raw e-mail
On Fri, Dec 3, 2010 at 1:25 PM, Peter Ankerstål <peter@pean.org> wrote: >>>> >>> >>> Yes, this is correct. But it still uses up 10MB from the parent >>> filesystem. This is a problem if you use both snapshots and quotas. >>> (and use alot of snapshots) >> >> You many be able to use the refquota attribute instead of quota unless >> you're trying to place quotas on filesystem hiearchies. >> >> -- Brooks > > Oh. Thanks. How could I miss that.. It seems I have to change some of the > "logic" in my setup for this to work as needed but overall this will work fine. If this is an NFS share, you might want to test what happens if you set a refquota, then exceed it. I know on OpenSolaris there have been some bugs where you get a 'no space left on device' when you try to rm files, in that situation, making it difficult for a user to recover from exceeding their quota. It's possible FreeBSD doesn't suffer from this, but you should test it if it's likely to come up.home | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?AANLkTinrmKj6kJ1N0-Ro6%2BkA04_KKRXCvdECT45zVyqA>