There are edge cases where non-ecc ram is ok with zfs... like if you aren't storing data there. My lab functional test system doesn't have ecc, my systems with my company data all have ecc. Test systems, or any place that has an entire copy of the exact same data, possibly in HPC. I have a customer that has rows(8 systems) of the exact same set of 100's of TB of data, but even then they use ecc. Just a mess to clean up later. So lab, functional test system to me is the only place it's ok. It is america though :) If non-ecc'ed ram passes zfs an inaccurate block.. It can be laid into the pool as bad metadata and make the whole pool unreadable. A couple bucks more for ecc ram and you have all your data!! linda On 11/20/14, 7:21 AM, T L wrote: > Andrew - > Thanks. My main box also has an AMD CPU, as it was the most cost > effective way to get ECC RAM. Hmmm -- speaking of which, I'd be > interested in your and Linda's thoughts on the necessity of ECC when > using ZFS. > > Thomas > > On Thursday, November 20, 2014, Andrew Berg > <aberg010 at my.hennepintech.edu <mailto:aberg010 at my.hennepintech.edu>> > wrote: > > On 2014.11.19 18:22, T L wrote: > > I'm very interested in bhyve. Please add it to the list. > My FreeBSD box has an AMD CPU and support for AMD's SVM just > landed in the > development branch a couple weeks ago, so I have not had a chance > to really use > it. I will talk a bit about it, but I can't do a demo, sorry. If > the bhyve AMD > patch gets MFC'd into 10-STABLE before the meeting, I will upgrade > to it and > play with bhyve a bit after the presentation, but it will be my > first time > really getting into it. > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org <javascript:;> > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mailman.mn-linux.org/pipermail/tclug-list/attachments/20141120/c4a17c7a/attachment.html>