From owner-freebsd-current Fri Sep 17 9: 0:58 1999 Delivered-To: freebsd-current@freebsd.org Received: from pau-amma.whistle.com (pau-amma.whistle.com [207.76.205.64]) by hub.freebsd.org (Postfix) with ESMTP id 5ABDF15504 for ; Fri, 17 Sep 1999 09:00:53 -0700 (PDT) (envelope-from dhw@whistle.com) Received: (from dhw@localhost) by pau-amma.whistle.com (8.9.2/8.9.2) id JAA08613; Fri, 17 Sep 1999 09:00:26 -0700 (PDT) Date: Fri, 17 Sep 1999 09:00:26 -0700 (PDT) From: David Wolfskill Message-Id: <199909171600.JAA08613@pau-amma.whistle.com> To: blk@skynet.be, dnelson@emsphone.com Subject: Re: More benchmarking stuff... Cc: freebsd-current@FreeBSD.ORG, tomdean@ix.netcom.com In-Reply-To: <19990917104608.A55059@dan.emsphone.com> Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG >Date: Fri, 17 Sep 1999 10:46:08 -0500 >From: Dan Nelson >Don't NetApps do logging, so if the system crashes, the files are >recovered from the log? It's my (ca. 4-year-ancient) recollection that the write requests are written to a split NVRAM buffer; when one half hits the high-water mark (or a timer pops, whichever happens first), the "active half" of the buffer is switched, and the half that had been written to has its entries committed to disk. So basically, yes -- the NetApp shouldn't lose data. (But "files" aren't written to the NVRAM; it's fairly raw write requests that get written.) Cheers, david -- David Wolfskill dhw@whistle.com UNIX System Administrator voice: (650) 577-7158 pager: (888) 347-0197 FAX: (650) 372-5915 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message