From owner-freebsd-virtualization@freebsd.org Fri Dec 1 09:54:03 2017 Return-Path: Delivered-To: freebsd-virtualization@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 3A373DFA36F for ; Fri, 1 Dec 2017 09:54:03 +0000 (UTC) (envelope-from FreeBSD@shaneware.biz) Received: from ipmail07.adl2.internode.on.net (ipmail07.adl2.internode.on.net [150.101.137.131]) by mx1.freebsd.org (Postfix) with ESMTP id 8F29C791A9; Fri, 1 Dec 2017 09:54:01 +0000 (UTC) (envelope-from FreeBSD@shaneware.biz) Received: from ppp121-45-11-215.bras1.adl4.internode.on.net (HELO leader.local) ([121.45.11.215]) by ipmail07.adl2.internode.on.net with ESMTP; 01 Dec 2017 20:23:53 +1030 Subject: Re: bhyve uses all available memory during IO-intensive operations To: Allan Jude , Dustin Wenz Cc: freebsd-virtualization@freebsd.org References: From: Shane Ambler Message-ID: Date: Fri, 1 Dec 2017 20:23:51 +1030 User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:52.0) Gecko/20100101 Thunderbird/52.4.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-AU Content-Transfer-Encoding: 8bit X-BeenThere: freebsd-virtualization@freebsd.org X-Mailman-Version: 2.1.25 Precedence: list List-Id: "Discussion of various virtualization techniques FreeBSD supports." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 01 Dec 2017 09:54:03 -0000 On 01/12/2017 13:43, Allan Jude wrote: > On 2017-11-30 22:10, Dustin Wenz wrote: >> I am using a zvol as the storage for the VM, and I do not have any ARC >> limits set. However, the bhyve process itself ends up grabbing the vast >> majority of memory.  >> >> I’ll run a test tomorrow to get the exact output from top. >> >>    - .Dustin >> >> On Nov 30, 2017, at 5:28 PM, Allan Jude > > wrote: >> >>> On 11/30/2017 18:15, Dustin Wenz wrote: >>>> I'm using chyves on FreeBSD 11.1 RELEASE to manage a few VMs (guest >>>> OS is also FreeBSD 11.1). Their sole purpose is to house some >>>> medium-sized Postgres databases (100-200GB). The host system has 64GB >>>> of real memory and 112GB of swap. I have configured each guest to >>>> only use 16GB of memory, yet while doing my initial database imports >>>> in the VMs, bhyve will quickly grow to use all available system >>>> memory and then be killed by the kernel: >>>> >>>>    kernel: swap_pager: I/O error - pageout failed; blkno 1735,size >>>> 4096, error 12 >>>>    kernel: swap_pager: I/O error - pageout failed; blkno 1610,size >>>> 4096, error 12 >>>>    kernel: swap_pager: I/O error - pageout failed; blkno 1763,size >>>> 4096, error 12 >>>>    kernel: pid 41123 (bhyve), uid 0, was killed: out of swap space That's the type of errors I see when wired jumps high. I'm not seeing this from bhyve but when your watching top, keep an eye on the wired amount. -- FreeBSD - the place to B...Sharing Devices Shane Ambler