From owner-freebsd-questions@FreeBSD.ORG Fri Oct 10 00:50:36 2014 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 1E918A45 for ; Fri, 10 Oct 2014 00:50:36 +0000 (UTC) Received: from mail.magehandbook.com (173-8-4-45-WashingtonDC.hfc.comcastbusiness.net [173.8.4.45]) by mx1.freebsd.org (Postfix) with ESMTP id E78296E5 for ; Fri, 10 Oct 2014 00:50:34 +0000 (UTC) Received: from [192.168.1.50] (Mac-Pro.magehandbook.com [192.168.1.50]) by mail.magehandbook.com (Postfix) with ESMTP id 3jDW0n5Fyfzmr for ; Thu, 9 Oct 2014 20:50:33 -0400 (EDT) Date: Thu, 09 Oct 2014 20:50:33 -0400 From: Daniel Staal Reply-To: FreeBSD Questions To: FreeBSD Questions Subject: Re: Next Steps to Debug ZFS Hang? Message-ID: <000F985485834C2758CC492A@[192.168.1.50]> In-Reply-To: <56695.76.192.184.214.1412896622.squirrel@cosmo.uchicago.edu> References: <1412732931033.813626ca@Nodemailer> <56695.76.192.184.214.1412896622.squirrel@cosmo.uchicago.edu> X-Mailer: Mulberry/4.0.8 (Mac OS X) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 10 Oct 2014 00:50:36 -0000 --As of October 9, 2014 6:17:02 PM -0500, Valeri Galtsev is alleged to have said: >> Note that I'm still just going by 'standard checks', but why don't you >> have >> any swap? I know you probably wouldn't use it with that much RAM around, >> but FreeBSD still performs better with it - and it wouldn't surprise me >> completely if that was causing your issue. > > I'm petrified. Is that so? I mean, as I understood you, 64 GB RAM machine > running under FreeBSD (say, 9.3) still needs some amount of SWAP for > better performance, right? --As for the rest, it is mine. Well, I suppose a better way to say it would be 'expects to have for normal operations'. Basically any modern OS expects to have some swap - even with abundant RAM. In theory you should be able to run without it, if you have the RAM, but it will be used (in small amounts) if it's available by normal housekeeping and operations. The tuning(7) man page under 9.3 recommends at least 256M of swap in all cases. (Though I note that recommendation is no longer in the man page under FreeBSD 10.) For a relatively new feature (like ZFS) that intensely and aggressively uses RAM (like ZFS), it would not surprise me if not having swap would uncover bugs in the implementation. Daniel T. Staal --------------------------------------------------------------- This email copyright the author. Unless otherwise noted, you are expressly allowed to retransmit, quote, or otherwise use the contents for non-commercial purposes. This copyright will expire 5 years after the author's death, or in 30 years, whichever is longer, unless such a period is in excess of local copyright law. ---------------------------------------------------------------