From owner-freebsd-performance@FreeBSD.ORG Sun Jan 18 23:12:15 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id AA43216A4CF for ; Sun, 18 Jan 2004 23:12:15 -0800 (PST) Received: from relay.pair.com (relay.pair.com [209.68.1.20]) by mx1.FreeBSD.org (Postfix) with SMTP id E44FC43D2D for ; Sun, 18 Jan 2004 23:12:10 -0800 (PST) (envelope-from silby@silby.com) Received: (qmail 59749 invoked from network); 19 Jan 2004 07:12:09 -0000 Received: from niwun.pair.com (HELO localhost) (209.68.2.70) by relay.pair.com with SMTP; 19 Jan 2004 07:12:09 -0000 X-pair-Authenticated: 209.68.2.70 Date: Mon, 19 Jan 2004 01:12:08 -0600 (CST) From: Mike Silbersack To: CHOI Junho In-Reply-To: <20040119.153452.10362034.cjh@kr.FreeBSD.org> Message-ID: <20040119010832.E85911@odysseus.silby.com> References: <20040119.153452.10362034.cjh@kr.FreeBSD.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: freebsd-net@freebsd.org cc: freebsd-performance@freebsd.org Subject: Re: mbuf tuning X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Jan 2004 07:12:15 -0000 On Mon, 19 Jan 2004, CHOI Junho wrote: > Hi, > > What is general guidelines of mbuf cluster tunables? I usually use There are no good guidelines other than "don't set it too high." Andre and I have talked about some ideas on how to make mbuf usage more dynamic, I think that he has something in the works. But at present, once you hit the wall, that's it. One way to reduce mbuf cluster usage is to use sendfile where possible. Data sent via sendfile does not use mbuf clusters, and is more memory efficient. If you run 5.2 or above, it's *much* more memory efficient, due to change Alan Cox recently made. Apache 2 will use sendfile by default, so if you're running apache 1, that may be one reason for an upgrade. > Increasing kern.ipc.nmbclusters caused frequent kernel panic > under 4.7/4.8/4.9. How can I set more nmbclusters value with 64K tcp > buffers? Or is any dependency for mbufclusters value? (e.g. RAM size, > kern.maxusers value or etc) > > p.s. RAM is 2G, Xeon 2.0G x 1 or 2 machines. You probably need to bump up KVA_PAGES to fit in all the extra mbuf clusters you're allocating. Mike "Silby" Silbersack From owner-freebsd-performance@FreeBSD.ORG Mon Jan 19 06:19:16 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 56F1F16A4CE; Mon, 19 Jan 2004 06:19:16 -0800 (PST) Received: from otter3.centtech.com (moat3.centtech.com [207.200.51.50]) by mx1.FreeBSD.org (Postfix) with ESMTP id AA3B943D3F; Mon, 19 Jan 2004 06:19:14 -0800 (PST) (envelope-from anderson@centtech.com) Received: from centtech.com (neutrino.centtech.com [10.177.171.220]) by otter3.centtech.com (8.12.3/8.12.3) with ESMTP id i0JEJD6T099222; Mon, 19 Jan 2004 08:19:13 -0600 (CST) (envelope-from anderson@centtech.com) Message-ID: <400BE749.2030009@centtech.com> Date: Mon, 19 Jan 2004 08:18:49 -0600 From: Eric Anderson User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Willem Jan Withagen References: <003c01c3de8d$d569edb0$471b3dd4@dual> In-Reply-To: <003c01c3de8d$d569edb0$471b3dd4@dual> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit cc: performance@freebsd.org cc: current@freebsd.org Subject: Re: Old SUN NFS performance papers. X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Jan 2004 14:19:16 -0000 Willem Jan Withagen wrote: >Hi, > >I had no responses to my recent question on the difference between NFS over UDP >and TCP. So perhaps nobody cares?? > >So I tried searching but have not found much yet. >Does anybody know where to find the white papers SUN once wrote about tuning >NFS??? They should be at sun, but where?? >All other suggestions to read are welcomed as well. > >Given my last posting I'm building two machines to do some NFS benchmark testing >on. >Suggestions on what people "always wanted to know (tm)" are also welcom, and >I'll see if I can get them integrated. >I've found the benchmarks in /usr/ports, some might do so nice work as well. > >If people are interested I'll keep them posted in performance@ > I'm definitely interested in what you find. I run a few heavily used FreeBSD NFS servers, and therefore always looking for tweaks and nobs to turn to make things better. In my experience, UDP has always been faster than TCP on NFS performance. Prior to 5.2, I have also seen mbuf related issues (all pretty much solvable with the right sysctl's). Let me know if I can help. Eric -- ------------------------------------------------------------------ Eric Anderson Systems Administrator Centaur Technology All generalizations are false, including this one. ------------------------------------------------------------------ From owner-freebsd-performance@FreeBSD.ORG Mon Jan 19 08:34:48 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D1B9D16A4CE for ; Mon, 19 Jan 2004 08:34:48 -0800 (PST) Received: from otter3.centtech.com (moat3.centtech.com [207.200.51.50]) by mx1.FreeBSD.org (Postfix) with ESMTP id 42DF843D31 for ; Mon, 19 Jan 2004 08:34:43 -0800 (PST) (envelope-from anderson@centtech.com) Received: from centtech.com (neutrino.centtech.com [10.177.171.220]) by otter3.centtech.com (8.12.3/8.12.3) with ESMTP id i0JGYe6T019983; Mon, 19 Jan 2004 10:34:41 -0600 (CST) (envelope-from anderson@centtech.com) Message-ID: <400C0707.7050805@centtech.com> Date: Mon, 19 Jan 2004 10:34:15 -0600 From: Eric Anderson User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Steve Francis References: <003c01c3de8d$d569edb0$471b3dd4@dual> <400BE749.2030009@centtech.com> <400C039B.6080403@expertcity.com> In-Reply-To: <400C039B.6080403@expertcity.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit cc: performance@freebsd.org cc: Willem Jan Withagen Subject: Re: Old SUN NFS performance papers. X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Jan 2004 16:34:48 -0000 Steve Francis wrote: > Benchmarking seems like the best thing to do, however I have some info > I've collected from prior posts: > > These are from the thread "Slow disk write speeds over network" on > freebsd-performance@freebsd.org, all written by Terry Lambert. > > you should definitely use TCP with FreeBSD NFS servers; it's > also just generally a good idea, since UDP frags act as a fixed > non-sliding window: NFS over UDP sucks. > > Also, you haven't said whether you are using aliases on your > network cards; aliases and NFS tend to interact badly. > > Finally, you probably want to tweak some sysctl's, e.g. > > net.inet.ip.check_interface=0 > net.inet.tcp.inflight_enable=1 > net.inet.tcp.inflight_debug=0 > net.inet.tcp.msl=3000 > net.inet.tcp.inflight_min=6100 > net.isr.enable=1 > > Given your overloading of your bus, that last one is probably > the most important one: it enables direct dispatch. > > You'll also want to enable DEVICE_POLLING in your kernel > config file (assuming you have a good ethernet card whose > driver supports it): > > options DEVICE_POLLING > options HZ=2000 > > > ...and yet more sysctl's for this: > > kern.polling.enable=1 > kern.polling.user_frac=50 # 0..100; whatever works best > > If you've got a really terrible Gigabit Ethernet card, then > you may be copying all your packets over again (e.g. m_pullup()), > and that could be eating your bus, too. > > >> Huh. I thought that the conventional wisdom was that on a local network >> with no packet loss (and therefore no re-transmission penalties), udp >> was >> way faster because the overhead was so much less. >> >> Sorry if this seems like a pretty basic question, but can you explain >> this? > > > Sure: > > 1) There is no such thing as no packet loss. > > 2) The UDP packets are reassembled in a reassembly queue > on the receiver. While this is happening, you can only > have one datagram outstanding at a time. With TCP, you > get a sliding window; with UDP, you stall waiting for > the reassembly, effectively giving you a non-sliding > window (request/response, with round trip latencies per > packet, instead of two of them amortized across a 100M > file transfer). > > 3) When a packet is lost, the UDP retransmit code is rather > crufty. It resends the whole series of packets, and you > eat the overhead for that. TCP, on the other hand, can > do selective acknowledgement, or, if it's not supported > by both ends, it can at least acknowledge the packets > that did get through, saving you a retransmit. > > 4) FreeBSD's UDP fragment reassembly buffer code is well > known to pretty much suck. This is true of most UDP > fragment reassembly code in the universe, however, and > is not that specific to FreeBSD. So sending UDP packets > that get fragged because they're larger than your MTU is > not a very clever way of achieving a fixed window size > larger than the MTU (see also #2, above, for why you do > not want to used an effectively fixed window protocol > anyway). > > Even if there were no packet loss at all with UDP, unless all > your data is around the size of one rsize/wsize/packet, the > combined RTT overhead for even a moderately large number of > packets in a single run is enough to trigger the amortized cost > of the additional TCP overhead being lower than the UDP overhead > from the latency. Depending on your hardware (switch latency, > half duplex, etc.), you could also be talking about a significant > combined bandwidth delay product. > > Now add to all this that you have to send explicit ACKs with UDP, > while you can use piggy-back ACKs on the return payloads for TCP. > > I think the idea that UDP was OK for nearly-lossless short-haul > came about from people who couldn't code a working TCP NFS client. > . > > > Eric Anderson wrote: > >> Willem Jan Withagen wrote: >> >>> Hi, >>> >>> I had no responses to my recent question on the difference between >>> NFS over UDP >>> and TCP. So perhaps nobody cares?? >>> >>> So I tried searching but have not found much yet. >>> Does anybody know where to find the white papers SUN once wrote >>> about tuning >>> NFS??? They should be at sun, but where?? >>> All other suggestions to read are welcomed as well. >>> >>> Given my last posting I'm building two machines to do some NFS >>> benchmark testing >>> on. >>> Suggestions on what people "always wanted to know (tm)" are also >>> welcom, and >>> I'll see if I can get them integrated. >>> I've found the benchmarks in /usr/ports, some might do so nice work >>> as well. >>> >>> If people are interested I'll keep them posted in performance@ >>> >> >> I'm definitely interested in what you find. I run a few heavily used >> FreeBSD NFS servers, and therefore always looking for tweaks and nobs >> to turn to make things better. In my experience, UDP has always been >> faster than TCP on NFS performance. Prior to 5.2, I have also seen >> mbuf related issues (all pretty much solvable with the right sysctl's). >> Let me know if I can help. >> >> Eric >> >> > I wasn't even sure where to start or stop snipping on this mail, since it is all good stuff - so I didn't. :) Thanks for the great info, and good explanations.. NFS+TCP is very nice, but I do believe the UDP transport was faster on a handful of tests (however I typically force use of TCP when I can).. One question - what does net.inet.ip.check_interface=0 do? Eric -- ------------------------------------------------------------------ Eric Anderson Systems Administrator Centaur Technology All generalizations are false, including this one. ------------------------------------------------------------------ From owner-freebsd-performance@FreeBSD.ORG Mon Jan 19 12:39:06 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8D2B316A4CE for ; Mon, 19 Jan 2004 12:39:06 -0800 (PST) Received: from otter3.centtech.com (moat3.centtech.com [207.200.51.50]) by mx1.FreeBSD.org (Postfix) with ESMTP id CBC3F43D5C for ; Mon, 19 Jan 2004 12:38:53 -0800 (PST) (envelope-from anderson@centtech.com) Received: from centtech.com (neutrino.centtech.com [10.177.171.220]) by otter3.centtech.com (8.12.3/8.12.3) with ESMTP id i0JKco6T064159; Mon, 19 Jan 2004 14:38:51 -0600 (CST) (envelope-from anderson@centtech.com) Message-ID: <400C403F.5000309@centtech.com> Date: Mon, 19 Jan 2004 14:38:23 -0600 From: Eric Anderson User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Steve Francis References: <003c01c3de8d$d569edb0$471b3dd4@dual> <400BE749.2030009@centtech.com> <400C039B.6080403@expertcity.com> <400C0707.7050805@centtech.com> <400C3D24.3080503@expertcity.com> In-Reply-To: <400C3D24.3080503@expertcity.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit cc: performance@freebsd.org cc: Willem Jan Withagen Subject: Re: Old SUN NFS performance papers. X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Jan 2004 20:39:06 -0000 Steve Francis wrote: > Eric Anderson wrote: > >> >> I wasn't even sure where to start or stop snipping on this mail, >> since it is all good stuff - so I didn't. :) Thanks for the great >> info, and good explanations.. NFS+TCP is very nice, but I do believe >> the UDP transport was faster on a handful of tests (however I >> typically force use of TCP when I can).. >> >> One question - what does net.inet.ip.check_interface=0 do? > > > > makes FreeBSD not care if the interface a response comes in on is the > same as the one a request did. Helps only if network topology is funky. That's handy for a network like I have. What would also be handy, is a sysctl like that for the client side - that tells FreeBSD to ignore the fact that a response is coming from a different IP than what it sent the request to. Yes, I know this is a security issue, and yes I understand the ramifications. Nevertheless, I need it - unless there is a way to tell redhat and solaris to always answer on the same interface the request came in on.. Eric -- ------------------------------------------------------------------ Eric Anderson Systems Administrator Centaur Technology All generalizations are false, including this one. ------------------------------------------------------------------ From owner-freebsd-performance@FreeBSD.ORG Mon Jan 19 12:58:15 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4E4E916A4CE for ; Mon, 19 Jan 2004 12:58:15 -0800 (PST) Received: from gray.impulse.net (gray.impulse.net [207.154.64.174]) by mx1.FreeBSD.org (Postfix) with ESMTP id C1E4D43D5F for ; Mon, 19 Jan 2004 12:58:10 -0800 (PST) (envelope-from ted@impulse.net) Received: by gray.impulse.net (Postfix, from userid 1000) id 7A5FFD6; Mon, 19 Jan 2004 12:58:10 -0800 (PST) To: Eric Anderson References: <003c01c3de8d$d569edb0$471b3dd4@dual> <400BE749.2030009@centtech.com> <400C039B.6080403@expertcity.com> <400C0707.7050805@centtech.com> <400C3D24.3080503@expertcity.com> <400C403F.5000309@centtech.com> From: Ted Cabeen Organization: Impulse Internet Services Date: Mon, 19 Jan 2004 12:58:10 -0800 In-Reply-To: <400C403F.5000309@centtech.com> (Eric Anderson's message of "Mon, 19 Jan 2004 14:38:23 -0600") Message-ID: <874qurhd0d.fsf@gray.impulse.net> User-Agent: Gnus/5.1002 (Gnus v5.10.2) XEmacs/21.4 (Reasonable Discussion, berkeley-unix) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii cc: Steve Francis cc: performance@freebsd.org cc: Willem Jan Withagen Subject: Re: Old SUN NFS performance papers. X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Jan 2004 20:58:15 -0000 Eric Anderson writes: > Steve Francis wrote: > >> Eric Anderson wrote: >> >>> >>> I wasn't even sure where to start or stop snipping on this mail, >>> since it is all good stuff - so I didn't. :) Thanks for the great >>> info, and good explanations.. NFS+TCP is very nice, but I do >>> believe the UDP transport was faster on a handful of tests (however >>> I typically force use of TCP when I can).. >>> >>> One question - what does net.inet.ip.check_interface=0 do? >> >> >> >> makes FreeBSD not care if the interface a response comes in on is >> the same as the one a request did. Helps only if network topology is >> funky. > > > That's handy for a network like I have. What would also be handy, is > a sysctl like that for the client side - that tells FreeBSD to ignore > the fact that a response is coming from a different IP than what it > sent the request to. Yes, I know this is a security issue, and yes I > understand the ramifications. Nevertheless, I need it - unless there > is a way to tell redhat and solaris to always answer on the same > interface the request came in on.. You can do that with policy routing on Linux. See the ip command. I don't know about Solaris. -- Ted Cabeen Sr. Systems/Network Administrator Impulse Internet Services From owner-freebsd-performance@FreeBSD.ORG Mon Jan 19 23:56:14 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 11D2516A4CE for ; Mon, 19 Jan 2004 23:56:14 -0800 (PST) Received: from relay.pair.com (relay.pair.com [209.68.1.20]) by mx1.FreeBSD.org (Postfix) with SMTP id 44C5943D53 for ; Mon, 19 Jan 2004 23:56:12 -0800 (PST) (envelope-from silby@silby.com) Received: (qmail 87331 invoked from network); 20 Jan 2004 07:56:10 -0000 Received: from niwun.pair.com (HELO localhost) (209.68.2.70) by relay.pair.com with SMTP; 20 Jan 2004 07:56:10 -0000 X-pair-Authenticated: 209.68.2.70 Date: Tue, 20 Jan 2004 01:56:09 -0600 (CST) From: Mike Silbersack To: CHOI Junho In-Reply-To: <20040119.192257.34695172.cjh@kr.FreeBSD.org> Message-ID: <20040120015356.N39477@odysseus.silby.com> References: <20040119.153452.10362034.cjh@kr.FreeBSD.org> <20040119.192257.34695172.cjh@kr.FreeBSD.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: freebsd-net@freebsd.org cc: freebsd-performance@freebsd.org Subject: Re: mbuf tuning X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 20 Jan 2004 07:56:14 -0000 On Mon, 19 Jan 2004, CHOI Junho wrote: > I am using custom version of thttpd. It allocates mmap() first(builtin > method of thttpd), and it try to use sendfile() if mmap() fails(out of > mmap memory). It really works good in normal status but the problem is > that sendfile buffer is also easy to flood. I need more sendfile > buffers but I don't know how to increase sendfile buffers either(I > think it's hidden sysctl but it was more difficult to tune than > nmbclusters). With higher traffic, thttpd sometimes stuck at "sfbufa" > status when I run top(I guess it's "sendfile buffer allocation" > status). > > 5.2 is fair good quality in my desktop but I have no experience in > production environment. I'll consider it once 5.x enters -STABLE tree, > but not now. > > Apache2 is one of my targets. How much better than apache-1.3.x in > static file service? thttpd using sendfile will certainly run circles around apache2, apache2 is still pre-fork. Under 4.x, you're going to have to tune the sfbufs by trial and error, but doing so will be worth it. Mike "Silby" Silbersack From owner-freebsd-performance@FreeBSD.ORG Tue Jan 20 12:56:32 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C7EA816A4CE for ; Tue, 20 Jan 2004 12:56:32 -0800 (PST) Received: from sizone.org (mortar.sizone.org [65.126.154.242]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0EA5143D6A for ; Tue, 20 Jan 2004 12:56:14 -0800 (PST) (envelope-from dgilbert@daveg.ca) Received: by sizone.org (Postfix, from userid 66) id 1D8EB30727; Tue, 20 Jan 2004 15:56:14 -0500 (EST) Received: by canoe.dclg.ca (Postfix, from userid 101) id 3D93E1D1E4B; Tue, 20 Jan 2004 15:56:12 -0500 (EST) From: David Gilbert MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <16397.38380.98493.182065@canoe.dclg.ca> Date: Tue, 20 Jan 2004 15:56:12 -0500 To: Eric Anderson In-Reply-To: <400C403F.5000309@centtech.com> References: <003c01c3de8d$d569edb0$471b3dd4@dual> <400BE749.2030009@centtech.com> <400C039B.6080403@expertcity.com> <400C0707.7050805@centtech.com> <400C3D24.3080503@expertcity.com> <400C403F.5000309@centtech.com> X-Mailer: VM 7.17 under 21.4 (patch 14) "Reasonable Discussion" XEmacs Lucid cc: Steve Francis cc: performance@freebsd.org cc: Willem Jan Withagen Subject: Re: Old SUN NFS performance papers. X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 20 Jan 2004 20:56:32 -0000 >>>>> "Eric" == Eric Anderson writes: Eric> Steve Francis wrote: >> Eric Anderson wrote: >> >>> I wasn't even sure where to start or stop snipping on this mail, >>> since it is all good stuff - so I didn't. :) Thanks for the great >>> info, and good explanations.. NFS+TCP is very nice, but I do >>> believe the UDP transport was faster on a handful of tests >>> (however I typically force use of TCP when I can).. >>> >>> One question - what does net.inet.ip.check_interface=0 do? >> makes FreeBSD not care if the interface a response comes in on is >> the same as the one a request did. Helps only if network topology >> is funky. Eric> That's handy for a network like I have. What would also be Eric> handy, is a sysctl like that for the client side - that tells Eric> FreeBSD to ignore the fact that a response is coming from a Eric> different IP than what it sent the request to. Yes, I know this Eric> is a security issue, and yes I understand the ramifications. Eric> Nevertheless, I need it - unless there is a way to tell redhat Eric> and solaris to always answer on the same interface the request Eric> came in on.. Another situation I've found it useful for is my laptop. As long as the IP it comes up on (via DHCP) is in exports, moving from one part of the lan to another works. It takes a minute or two, but the laptop will reopen the TCP connection and the mount comes back alive. Dave. -- ============================================================================ |David Gilbert, Independent Contractor. | Two things can only be | |Mail: dave@daveg.ca | equal if and only if they | |http://daveg.ca | are precisely opposite. | =========================================================GLO================ From owner-freebsd-performance@FreeBSD.ORG Tue Jan 13 14:12:40 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1915916A4CF for ; Tue, 13 Jan 2004 14:12:40 -0800 (PST) Received: from wind.mindcry.org (nat-gr.wmis.net [216.109.194.252]) by mx1.FreeBSD.org (Postfix) with ESMTP id 371DF43D58 for ; Tue, 13 Jan 2004 14:10:43 -0800 (PST) (envelope-from david@wind.mindcry.org) Received: by wind.mindcry.org (Postfix, from userid 1001) id A7DCB417B; Tue, 13 Jan 2004 17:10:42 -0500 (EST) Date: Tue, 13 Jan 2004 17:10:42 -0500 From: David Hill To: Sean Chittenden Message-ID: <20040113221042.GA3616@phobia.ms> References: <20040113143650.GA1424@phobia.ms> <81F83F77-4612-11D8-A8F0-000A95C705DC@chittenden.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <81F83F77-4612-11D8-A8F0-000A95C705DC@chittenden.org> User-Agent: Mutt/1.5.5.1i X-Mailman-Approved-At: Fri, 23 Jan 2004 00:09:07 -0800 cc: performance@freebsd.org Subject: Re: postgresql on freebsd - lots of connections X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 13 Jan 2004 22:12:40 -0000 On Tue, Jan 13, 2004 at 01:50:31PM -0800, Sean Chittenden wrote: > [ Howdy. My network's reverse DNS is fubar'ed at the moment, can you > CC either ] > [ database@FreeBSD.org or performance@FreeBSD.org when you reply? I > think ] > [ others might like to either listen, or contribute to this discussion. > ] > > >Sean - > >If i can borrow your brain for a few :) > > > >I am running postfix, postgresql, and courier together. postgresql > >needs to be > >fast for reading. > > SELECT happy, eh? > > >I dont need to be able to support huge queriesor results.. they are > >all "SELECT blah FROM table WHERE domain='domain.com'" > >The tables are unique'd index'd. > > Good to hear, UNIQUE INDEXes are faster than non-unique INDEXes. You > don't have control over the SQL such that you can make the various > programs use persistent connections and/or prepared statements, do you? > > >I am running postgresql with both postfix and courier querying it for > >aliases, users, and relay_domains... mail is not stored in SQL. > > The biggest factor in speeding things up will be persistent connections. > > >There will be about 4 postfix servers and 2 or 3 courier servers (we > >have a large email userbase) accessing the postgresql server when > >finished. > > > >What are some good FreeBSD kernel tuning options and postgresql tuning > >options to support a large number of connections returning very small > >results? > > Hrm.... well, as stated, anything you can do to reduce the connection > startup time is going to be key. If you want to quasi-hack a custom > version of PostgreSQL and connections aren't being cached, you'd > probably want to have the _client_ do something like: > > int optval = 1; > setsockopt(s, SOL_TCP, TCP_NOPUSH, &optval, sizeof(optval)); > > Actually, here's the patch to make this happen (also at > http://people.FreeBSD.org/~seanc/patches/#pgsql-tcp_nopush). I haven't > tested the performance impact of this and I don't know if this will > impact interactive sessions or not, but, I'd hope that it'll speed > things up and reduce the packet flow since now the server shouldn't > flush the socket after every row.. which could cause a startup delay, > but when it comes to sending data and closing the connection, it should > be a win. *shrug* Someone with more TTCP foo than me may be able to > predict better than I. With HTTP, small requests can be handled in > three packets, so who knows. I'd be interested in any impact you > notice with this. So let's see... what else can be done. > > > > Setting net.inet.tcp.delayed_ack=1 would be a good idea probably, > reduces the number of TCP packets. Beyond that, there's not a whole > lot that you can do other than possibly preloading plpgsql.so if you > make use of that. Other things that you may want to _test_ heavily, > would be futzing around with the block size. Only the -devel port has > this option, but you may find that SELECTs will be faster at 4K than 8K > or 16K. It's hard to say though... if you increase the caching and are > able to keep the entire database in the OS's cache (you may want to > increase the amount of kernel space available for that, NBUF default * > 2 && and BKVASIZE default * 4), it may be to your advantage to > _increase_ the block size to something larger like 16K or even possibly > 32K, though be sure to change your postgresql.conf settings when you > tweak the page size). > > ... and that's about all I can think of now. Let me know how your > testing goes though as this is something that I'm going to need to > spend some time working on later this month (*smells libevent + > PostgreSQL coming real soon*). -sc > > -- > Sean Chittenden > seanc@FreeBSD.org > http://people.FreeBSD.org/~seanc/ I have Postfix using proxymap to (share one open table among multiple processes), which helped a lot. My network memory buffers are fine, hardly being used. I pulled some of your FreeBSD kernel options from the postgresql performance mail-list to get mine handling about 128 connections. The most I have seen postgresql need to open so far is 45, so I might be okay. I have 2 postfix servers (2.4Ghz celeron, 512MB ram, 40GB IDE, 3com NIC's) accessing postgresql (same hardware config). Load is about 0.7, a lot of inact/free ram, and network bugs are at a minimum. # freebsd kernel options options SHMMAXPGS=65536 options SEMMNI=40 options SEMMNS=240 options SEMUME=40 options SEMMNU=120 # postgresql.conf options max_connections = 128 shared_buffers = 2048 effective_cache_size = (sysctl -n vfs.hibufspace / 8192) - David From owner-freebsd-performance@FreeBSD.ORG Sun Jan 18 22:36:17 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7464A16A4CE; Sun, 18 Jan 2004 22:36:17 -0800 (PST) Received: from daemon.kr.FreeBSD.org (daemon.kr.freebsd.org [61.78.53.31]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6BE5B43D2F; Sun, 18 Jan 2004 22:36:15 -0800 (PST) (envelope-from cjh@kr.FreeBSD.org) Received: from localhost (gradius [211.44.63.164]) by daemon.kr.FreeBSD.org (Postfix) with ESMTP id 4FD121A744; Mon, 19 Jan 2004 15:34:50 +0900 (KST) Date: Mon, 19 Jan 2004 15:34:52 +0900 (KST) Message-Id: <20040119.153452.10362034.cjh@kr.FreeBSD.org> To: freebsd-net@freebsd.org From: CHOI Junho Organization: Korea FreeBSD Users Group X-URL: http://www.kr.FreeBSD.org/~cjh X-Mailer: Mew version 4.0.62 on Emacs 21.3.1 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Fri, 23 Jan 2004 00:09:07 -0800 cc: freebsd-performance@freebsd.org Subject: mbuf tuning X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Jan 2004 06:36:17 -0000 Hi, What is general guidelines of mbuf cluster tunables? I usually use kern.ipc.nmbclusters="65536" in /boot/loader.conf.local. But it has limits on concurrent TCP sessions, under my /etc/sysctl.conf configuration: net.inet.tcp.sendspace=65536 net.inet.tcp.recvspace=65536 With above parameters, about 1000 connection was the maximum. Of course if I set 1/2 value for tcp buffer(32768), maximum connection is doubled. But I want more. Increasing kern.ipc.nmbclusters caused frequent kernel panic under 4.7/4.8/4.9. How can I set more nmbclusters value with 64K tcp buffers? Or is any dependency for mbufclusters value? (e.g. RAM size, kern.maxusers value or etc) p.s. RAM is 2G, Xeon 2.0G x 1 or 2 machines. -- CHOI Junho KFUG FreeBSD Project Web Data Bank Key fingerprint = 1369 7374 A45F F41A F3C0 07E3 4A01 C020 E602 60F5 From owner-freebsd-performance@FreeBSD.ORG Mon Jan 19 02:24:21 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E5B9816A4CE; Mon, 19 Jan 2004 02:24:21 -0800 (PST) Received: from daemon.kr.FreeBSD.org (daemon.kr.freebsd.org [61.78.53.31]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2910F43D49; Mon, 19 Jan 2004 02:24:20 -0800 (PST) (envelope-from cjh@kr.FreeBSD.org) Received: from localhost (gradius [211.44.63.164]) by daemon.kr.FreeBSD.org (Postfix) with ESMTP id 87F631A6EB; Mon, 19 Jan 2004 19:22:53 +0900 (KST) Date: Mon, 19 Jan 2004 19:22:57 +0900 (KST) Message-Id: <20040119.192257.34695172.cjh@kr.FreeBSD.org> To: silby@silby.com From: CHOI Junho In-Reply-To: <20040119010832.E85911@odysseus.silby.com> References: <20040119.153452.10362034.cjh@kr.FreeBSD.org> <20040119010832.E85911@odysseus.silby.com> Organization: Korea FreeBSD Users Group X-URL: http://www.kr.FreeBSD.org/~cjh X-Mailer: Mew version 4.0.62 on Emacs 21.3.1 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Fri, 23 Jan 2004 00:09:07 -0800 cc: freebsd-net@freebsd.org cc: freebsd-performance@freebsd.org Subject: Re: mbuf tuning X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Jan 2004 10:24:22 -0000 From: Mike Silbersack Subject: Re: mbuf tuning Date: Mon, 19 Jan 2004 01:12:08 -0600 (CST) > There are no good guidelines other than "don't set it too high." Andre > and I have talked about some ideas on how to make mbuf usage more dynamic, > I think that he has something in the works. But at present, once you hit > the wall, that's it. > > One way to reduce mbuf cluster usage is to use sendfile where possible. > Data sent via sendfile does not use mbuf clusters, and is more memory > efficient. If you run 5.2 or above, it's *much* more memory efficient, > due to change Alan Cox recently made. Apache 2 will use sendfile by > default, so if you're running apache 1, that may be one reason for an > upgrade. I am using custom version of thttpd. It allocates mmap() first(builtin method of thttpd), and it try to use sendfile() if mmap() fails(out of mmap memory). It really works good in normal status but the problem is that sendfile buffer is also easy to flood. I need more sendfile buffers but I don't know how to increase sendfile buffers either(I think it's hidden sysctl but it was more difficult to tune than nmbclusters). With higher traffic, thttpd sometimes stuck at "sfbufa" status when I run top(I guess it's "sendfile buffer allocation" status). 5.2 is fair good quality in my desktop but I have no experience in production environment. I'll consider it once 5.x enters -STABLE tree, but not now. Apache2 is one of my targets. How much better than apache-1.3.x in static file service? > > Increasing kern.ipc.nmbclusters caused frequent kernel panic > > under 4.7/4.8/4.9. How can I set more nmbclusters value with 64K tcp > > buffers? Or is any dependency for mbufclusters value? (e.g. RAM size, > > kern.maxusers value or etc) > > > > p.s. RAM is 2G, Xeon 2.0G x 1 or 2 machines. > > You probably need to bump up KVA_PAGES to fit in all the extra mbuf > clusters you're allocating. Can you tell me in more detail? > Mike "Silby" Silbersack Thanks, -- CHOI Junho KFUG FreeBSD Project Web Data Bank Key fingerprint = 1369 7374 A45F F41A F3C0 07E3 4A01 C020 E602 60F5 From owner-freebsd-performance@FreeBSD.ORG Mon Jan 19 02:33:56 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2156B16A4CE; Mon, 19 Jan 2004 02:33:56 -0800 (PST) Received: from relay.macomnet.ru (relay.macomnet.ru [195.128.64.10]) by mx1.FreeBSD.org (Postfix) with ESMTP id 73A1E43D4C; Mon, 19 Jan 2004 02:33:53 -0800 (PST) (envelope-from maxim@macomnet.ru) Received: from news1.macomnet.ru (gog401zp@news1.macomnet.ru [195.128.64.14]) by relay.macomnet.ru (8.12.10/8.12.10) with ESMTP id i0JAWGhg3588493; Mon, 19 Jan 2004 13:32:16 +0300 (MSK) Date: Mon, 19 Jan 2004 13:32:16 +0300 (MSK) From: Maxim Konovalov To: CHOI Junho In-Reply-To: <20040119.192257.34695172.cjh@kr.FreeBSD.org> Message-ID: <20040119132741.O42397@news1.macomnet.ru> References: <20040119.153452.10362034.cjh@kr.FreeBSD.org> <20040119.192257.34695172.cjh@kr.FreeBSD.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Mailman-Approved-At: Fri, 23 Jan 2004 00:09:07 -0800 cc: freebsd-net@freebsd.org cc: silby@silby.com cc: freebsd-performance@freebsd.org Subject: Re: mbuf tuning X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Jan 2004 10:33:56 -0000 On Mon, 19 Jan 2004, 19:22+0900, CHOI Junho wrote: > From: Mike Silbersack > Subject: Re: mbuf tuning > Date: Mon, 19 Jan 2004 01:12:08 -0600 (CST) > > > There are no good guidelines other than "don't set it too high." Andre > > and I have talked about some ideas on how to make mbuf usage more dynamic, > > I think that he has something in the works. But at present, once you hit > > the wall, that's it. > > > > One way to reduce mbuf cluster usage is to use sendfile where possible. > > Data sent via sendfile does not use mbuf clusters, and is more memory > > efficient. If you run 5.2 or above, it's *much* more memory efficient, > > due to change Alan Cox recently made. Apache 2 will use sendfile by > > default, so if you're running apache 1, that may be one reason for an > > upgrade. > > I am using custom version of thttpd. It allocates mmap() first(builtin > method of thttpd), and it try to use sendfile() if mmap() fails(out of > mmap memory). It really works good in normal status but the problem is > that sendfile buffer is also easy to flood. I need more sendfile > buffers but I don't know how to increase sendfile buffers either(I > think it's hidden sysctl but it was more difficult to tune than > nmbclusters). With higher traffic, thttpd sometimes stuck at "sfbufa" > status when I run top(I guess it's "sendfile buffer allocation" > status). man 2 sendfile, man 7 tuning are a good start. In 5.2 you can monitor sendfile buffers usage via kern.ipc.nsfbufs* sysctls or netstat(1). [...] -- Maxim Konovalov, maxim@macomnet.ru, maxim@FreeBSD.org From owner-freebsd-performance@FreeBSD.ORG Mon Jan 19 05:12:23 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D14A916A4CE; Mon, 19 Jan 2004 05:12:23 -0800 (PST) Received: from freebee.digiware.nl (dsl144.iae.nl [212.61.62.145]) by mx1.FreeBSD.org (Postfix) with ESMTP id 72F4C43D1D; Mon, 19 Jan 2004 05:12:21 -0800 (PST) (envelope-from wjw@withagen.nl) Received: from dual (dual [212.61.27.71]) by freebee.digiware.nl (8.12.10/8.12.9) with SMTP id i0JDDWeL087623; Mon, 19 Jan 2004 14:13:33 +0100 (CET) (envelope-from wjw@withagen.nl) Message-ID: <003c01c3de8d$d569edb0$471b3dd4@dual> From: "Willem Jan Withagen" To: , Date: Mon, 19 Jan 2004 14:12:04 +0100 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1158 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165 X-Mailman-Approved-At: Fri, 23 Jan 2004 00:09:07 -0800 Subject: Old SUN NFS performance papers. X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Jan 2004 13:12:24 -0000 Hi, I had no responses to my recent question on the difference between NFS over UDP and TCP. So perhaps nobody cares?? So I tried searching but have not found much yet. Does anybody know where to find the white papers SUN once wrote about tuning NFS??? They should be at sun, but where?? All other suggestions to read are welcomed as well. Given my last posting I'm building two machines to do some NFS benchmark testing on. Suggestions on what people "always wanted to know (tm)" are also welcom, and I'll see if I can get them integrated. I've found the benchmarks in /usr/ports, some might do so nice work as well. If people are interested I'll keep them posted in performance@ --WjW From owner-freebsd-performance@FreeBSD.ORG Mon Jan 19 13:20:08 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9358E16A4CE; Mon, 19 Jan 2004 13:20:08 -0800 (PST) Received: from park.rambler.ru (park.rambler.ru [81.19.64.101]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6BB0443D54; Mon, 19 Jan 2004 13:20:01 -0800 (PST) (envelope-from is@rambler-co.ru) Received: from is (is.park.rambler.ru [81.19.64.102]) by park.rambler.ru (8.12.6/8.12.6) with ESMTP id i0JLJsAY009298; Tue, 20 Jan 2004 00:19:54 +0300 (MSK) (envelope-from is@rambler-co.ru) Date: Tue, 20 Jan 2004 00:19:54 +0300 (MSK) From: Igor Sysoev X-Sender: is@is To: CHOI Junho In-Reply-To: <20040119.192257.34695172.cjh@kr.FreeBSD.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Mailman-Approved-At: Fri, 23 Jan 2004 00:09:07 -0800 cc: freebsd-net@freebsd.org cc: freebsd-performance@freebsd.org Subject: Re: mbuf tuning X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Jan 2004 21:20:08 -0000 On Mon, 19 Jan 2004, CHOI Junho wrote: > From: Mike Silbersack > Subject: Re: mbuf tuning > Date: Mon, 19 Jan 2004 01:12:08 -0600 (CST) > > > There are no good guidelines other than "don't set it too high." Andre > > and I have talked about some ideas on how to make mbuf usage more dynamic, > > I think that he has something in the works. But at present, once you hit > > the wall, that's it. > > > > One way to reduce mbuf cluster usage is to use sendfile where possible. > > Data sent via sendfile does not use mbuf clusters, and is more memory > > efficient. If you run 5.2 or above, it's *much* more memory efficient, > > due to change Alan Cox recently made. Apache 2 will use sendfile by > > default, so if you're running apache 1, that may be one reason for an > > upgrade. > > I am using custom version of thttpd. It allocates mmap() first(builtin > method of thttpd), and it try to use sendfile() if mmap() fails(out of > mmap memory). It really works good in normal status but the problem is > that sendfile buffer is also easy to flood. I need more sendfile > buffers but I don't know how to increase sendfile buffers either(I > think it's hidden sysctl but it was more difficult to tune than > nmbclusters). With higher traffic, thttpd sometimes stuck at "sfbufa" > status when I run top(I guess it's "sendfile buffer allocation" > status). In 4.x you have to rebuild the kernel with options NSFBUFS=16384 It equals to (512 + maxusers * 16) by default. By the way, why do you want to use the big net.inet.tcp.sendspace and net.inet.tcp.recvspace ? It makes a sense for Apache but thttpd can easy work with the small buffers, say, 16K or even 8K. > > > Increasing kern.ipc.nmbclusters caused frequent kernel panic > > > under 4.7/4.8/4.9. How can I set more nmbclusters value with 64K tcp > > > buffers? Or is any dependency for mbufclusters value? (e.g. RAM size, > > > kern.maxusers value or etc) > > > > > > p.s. RAM is 2G, Xeon 2.0G x 1 or 2 machines. > > > > You probably need to bump up KVA_PAGES to fit in all the extra mbuf > > clusters you're allocating. > > Can you tell me in more detail? >From LINT: --- # # Change the size of the kernel virtual address space. Due to # constraints in loader(8) on i386, this must be a multiple of 4. # 256 = 1 GB of kernel address space. Increasing this also causes # a reduction of the address space in user processes. 512 splits # the 4GB cpu address space in half (2GB user, 2GB kernel). # options KVA_PAGES=260 --- Default KVA_PAGES are 256. Igor Sysoev http://sysoev.ru/en/ From owner-freebsd-performance@FreeBSD.ORG Fri Jan 23 11:20:32 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 97A2616A4CE for ; Fri, 23 Jan 2004 11:20:32 -0800 (PST) Received: from hotmail.com (sea2-dav53.sea2.hotmail.com [207.68.164.46]) by mx1.FreeBSD.org (Postfix) with ESMTP id 81D4243D99 for ; Fri, 23 Jan 2004 11:18:14 -0800 (PST) (envelope-from forharryh@hotmail.com) Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC; Fri, 23 Jan 2004 11:14:48 -0800 Received: from 66.81.126.223 by sea2-dav53.sea2.hotmail.com with DAV; Fri, 23 Jan 2004 19:14:48 +0000 X-Originating-IP: [66.81.126.223] X-Originating-Email: [forharryh@hotmail.com] X-Sender: forharryh@hotmail.com From: "HarryH" To: Date: Fri, 23 Jan 2004 11:14:48 -0800 MIME-Version: 1.0 X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1158 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165 Message-ID: X-OriginalArrivalTime: 23 Jan 2004 19:14:48.0168 (UTC) FILETIME=[2AFFE680:01C3E1E5] Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.1 Subject: FreeBSD & Solais 9 questions X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 23 Jan 2004 19:20:32 -0000 Hi, I have BSD 4.8 and want to install it on a server that already had = Windows 2000 Server and Advanced Server. I want to install BSD on a = spare partition so my questions are: 1. Will BSD install for multiple boot with Windows already there and = if so,=20 2. How much disk space do I need for a good server/workstation = install - no games or source code - will 5GB do the job? Do I need = more? 3. What are the potential problems that I need to be aware of during = and after installation? I just installed Solaris 9 X86 on an older PC with a 4GB HD, and it will = not allow me to logon. It issues a message, "DT Messaging system could = not be started" and tells me how to work around it with a "failsafe = logon" and files to check. I have been all over the Internet looking = for solutions. This appears to be a fairly common problem but none of = the solutions seem to really work. Do any of you know of a good Solaris = Users Group? I am new to Unix. Thanks, Harry =20 From owner-freebsd-performance@FreeBSD.ORG Fri Jan 23 11:36:20 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2BA5816A4CE for ; Fri, 23 Jan 2004 11:36:20 -0800 (PST) Received: from smtpout.mac.com (smtpout.mac.com [17.250.248.86]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1AE6643D46 for ; Fri, 23 Jan 2004 11:36:18 -0800 (PST) (envelope-from cswiger@mac.com) Received: from mac.com (smtpin07-en2 [10.13.10.152]) by smtpout.mac.com (Xserve/MantshX 2.0) with ESMTP id i0NJaHBQ007418; Fri, 23 Jan 2004 11:36:17 -0800 (PST) Received: from [10.1.1.193] (nfw2.codefab.com [66.234.138.66]) (authenticated bits=0) by mac.com (Xserve/smtpin07/MantshX 3.0) with ESMTP id i0NJZxeO024574; Fri, 23 Jan 2004 11:36:16 -0800 (PST) In-Reply-To: References: Mime-Version: 1.0 (Apple Message framework v609) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <5EEFD610-4DDB-11D8-A782-003065ABFD92@mac.com> Content-Transfer-Encoding: 7bit From: Charles Swiger Date: Fri, 23 Jan 2004 14:35:59 -0500 To: HarryH X-Mailer: Apple Mail (2.609) cc: freebsd-performance@freebsd.org Subject: Re: FreeBSD & Solais 9 questions X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 23 Jan 2004 19:36:20 -0000 On Jan 23, 2004, at 2:14 PM, HarryH wrote: > I have BSD 4.8 and want to install it on a server that already had > Windows 2000 Server and Advanced Server. I want to install BSD on a > spare partition so my questions are: > 1. Will BSD install for multiple boot with Windows already there > and if so, Yes, the FreeBSD bootmanager will let you choose to boot Windows partitions as well. Or you can configure Win2K with an OS menu to boot FreeBSD, if you wanted. > 2. How much disk space do I need for a good server/workstation > install - no games or source code - will 5GB do the job? Do I need > more? 5GB is more than enough for the entire OS and sources; but how much you use depends on what you're doing, of course. > 3. What are the potential problems that I need to be aware of > during and after installation? It's a good idea to backup any data you care about before the install, obviously. > I just installed Solaris 9 X86 on an older PC with a 4GB HD, and it > will not allow me to logon. It issues a message, "DT Messaging system > could not be started" and tells me how to work around it with a > "failsafe logon" and files to check. I have been all over the > Internet looking for solutions. This appears to be a fairly common > problem but none of the solutions seem to really work. Do any of you > know of a good Solaris Users Group? Off-topic here; try a comp.sys.sun.* newsgroup, perhaps. -- -Chuck From owner-freebsd-performance@FreeBSD.ORG Fri Jan 23 22:12:08 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2EAE116A4CE for ; Fri, 23 Jan 2004 22:12:08 -0800 (PST) Received: from hotmail.com (sea2-dav55.sea2.hotmail.com [207.68.164.63]) by mx1.FreeBSD.org (Postfix) with ESMTP id 89FA443D2D for ; Fri, 23 Jan 2004 22:10:34 -0800 (PST) (envelope-from forharryh@hotmail.com) Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC; Fri, 23 Jan 2004 22:08:19 -0800 Received: from 66.81.123.218 by sea2-dav55.sea2.hotmail.com with DAV; Sat, 24 Jan 2004 06:08:19 +0000 X-Originating-IP: [66.81.123.218] X-Originating-Email: [forharryh@hotmail.com] X-Sender: forharryh@hotmail.com From: "HarryH" To: Date: Fri, 23 Jan 2004 22:08:25 -0800 MIME-Version: 1.0 X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1158 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165 Message-ID: X-OriginalArrivalTime: 24 Jan 2004 06:08:19.0986 (UTC) FILETIME=[77107F20:01C3E240] Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.1 Subject: A good BSD Text Book? X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 24 Jan 2004 06:12:08 -0000 Hi, Can anyone recommend a/some really good BSD (4.8) books/manuals, for not = only a BSD beginner, but for someone that will really get into detail in = a short time? What I found at Border's was a real beginner's book. Or, = will I end up with a couple of books? I have the "Unix in a Nutshell" = by O'Reilly but would like to zero in on BSD. Thanks, Harry From owner-freebsd-performance@FreeBSD.ORG Fri Jan 23 22:38:08 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id CF2B316A4CE for ; Fri, 23 Jan 2004 22:38:08 -0800 (PST) Received: from santiago.pacific.net.sg (santiago.pacific.net.sg [203.120.90.135]) by mx1.FreeBSD.org (Postfix) with SMTP id B7B7F43D49 for ; Fri, 23 Jan 2004 22:38:06 -0800 (PST) (envelope-from oceanare@pacific.net.sg) Received: (qmail 24335 invoked from network); 24 Jan 2004 06:38:05 -0000 Received: from unknown (HELO maxwell2.pacific.net.sg) (203.120.90.192) by santiago with SMTP; 24 Jan 2004 06:38:04 -0000 Received: from pacific.net.sg ([210.24.202.41]) by maxwell2.pacific.net.sg with ESMTP id <20040124063804.UIBZ2304.maxwell2.pacific.net.sg@pacific.net.sg>; Sat, 24 Jan 2004 14:38:04 +0800 Message-ID: <401212CC.4090005@pacific.net.sg> Date: Sat, 24 Jan 2004 14:38:04 +0800 From: Erich Dollansky Organization: oceanare pte ltd User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.6) Gecko/20040113 X-Accept-Language: en, en-us, de MIME-Version: 1.0 To: HarryH References: In-Reply-To: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: freebsd-performance@freebsd.org Subject: Re: A good BSD Text Book? X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 24 Jan 2004 06:38:08 -0000 Hi, I have FreeBSD unleashed written by Micheal Urban and Brian Tiemann published by SAMS. I would not say that it goes very deep into it but gives the hints you need to be able to dig deeper. My copy is based on earlier 4.x and 5.0 version of FreeBSD. It should be updated meanwhile. I would not say it is really good, but it was good enough for me to start with FreeBSD those days. Erich HarryH wrote: > Hi, > Can anyone recommend a/some really good BSD (4.8) books/manuals, for not only a BSD beginner, but for someone that will really get into detail in a short time? What I found at Border's was a real beginner's book. Or, will I end up with a couple of books? I have the "Unix in a Nutshell" by O'Reilly but would like to zero in on BSD. > > Thanks, > Harry > _______________________________________________ > freebsd-performance@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-performance > To unsubscribe, send any mail to "freebsd-performance-unsubscribe@freebsd.org" > From owner-freebsd-performance@FreeBSD.ORG Sat Jan 24 16:17:50 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id B6FFA16A4CE for ; Sat, 24 Jan 2004 16:17:50 -0800 (PST) Received: from 82-41-27-158.cable.ubr04.edin.blueyonder.co.uk (82-41-27-158.cable.ubr04.edin.blueyonder.co.uk [82.41.27.158]) by mx1.FreeBSD.org (Postfix) with ESMTP id D599F43D31 for ; Sat, 24 Jan 2004 16:17:47 -0800 (PST) (envelope-from andrew@cream.org) Received: from cream.org (spatula.flat [192.168.0.2]) by myriad.flat (Postfix) with ESMTP id ACDE4E4; Sat, 24 Jan 2004 23:09:50 +0000 (GMT) Message-ID: <40130B3C.90703@cream.org> Date: Sun, 25 Jan 2004 00:18:04 +0000 From: Andrew Boothman User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.6b) Gecko/20031205 Thunderbird/0.4 X-Accept-Language: en-us, en MIME-Version: 1.0 To: HarryH References: In-Reply-To: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: freebsd-performance@freebsd.org Subject: Re: A good BSD Text Book? X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 25 Jan 2004 00:17:50 -0000 HarryH wrote: > Hi, > Can anyone recommend a/some really good BSD (4.8) books/manuals, for not only a BSD beginner, but for someone that will really get into detail in a short time? What I found at Border's was a real beginner's book. Or, will I end up with a couple of books? I have the "Unix in a Nutshell" by O'Reilly but would like to zero in on BSD. Absolute BSD by Michael Lucas is a good read and gets into some good detail. Also check out The Complete FreeBSD by Greg Lehey (4th Ed.) is also very good. Andrew From owner-freebsd-performance@FreeBSD.ORG Sat Jan 24 18:17:03 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 81AF816A4CE; Sat, 24 Jan 2004 18:17:03 -0800 (PST) Received: from fledge.watson.org (fledge.watson.org [204.156.12.50]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8B02443D1F; Sat, 24 Jan 2004 18:17:01 -0800 (PST) (envelope-from robert@fledge.watson.org) Received: from fledge.watson.org (localhost [127.0.0.1]) by fledge.watson.org (8.12.10/8.12.10) with ESMTP id i0P2EqUd069682; Sat, 24 Jan 2004 21:14:52 -0500 (EST) (envelope-from robert@fledge.watson.org) Received: from localhost (robert@localhost)i0P2EptA069679; Sat, 24 Jan 2004 21:14:52 -0500 (EST) (envelope-from robert@fledge.watson.org) Date: Sat, 24 Jan 2004 21:14:51 -0500 (EST) From: Robert Watson X-Sender: robert@fledge.watson.org To: Willem Jan Withagen In-Reply-To: <003c01c3de8d$d569edb0$471b3dd4@dual> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: performance@FreeBSD.ORG cc: current@FreeBSD.ORG Subject: Re: Old SUN NFS performance papers. X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 25 Jan 2004 02:17:03 -0000 On Mon, 19 Jan 2004, Willem Jan Withagen wrote: > I had no responses to my recent question on the difference between NFS > over UDP and TCP. So perhaps nobody cares?? > > So I tried searching but have not found much yet. Does anybody know > where to find the white papers SUN once wrote about tuning NFS??? They > should be at sun, but where?? All other suggestions to read are > welcomed as well. > > Given my last posting I'm building two machines to do some NFS benchmark > testing on. Suggestions on what people "always wanted to know (tm)" are > also welcom, and I'll see if I can get them integrated. I've found the > benchmarks in /usr/ports, some might do so nice work as well. > > If people are interested I'll keep them posted in performance@ --WjW I haven't done much benchmarking on NFS lately, but something worth remembering is that people have spent a lot of time researching and optimizing TCP for a variety of connection types, whereas the NFS code basically has a static implementation of RPC backoff and flow control that hasn't evolved much. TCP is aware of things like the pathwise-mtu to the server and adapts, whereas UDP just loses packets due to fragmentation, especially if you are using larger block sizes. Please do post your discoveries on performance@, and perhaps we could build an NFS performance tuning section in the FreeBSD Handbook (or if there's not that much content, add it to the FAQ)? Robert N M Watson FreeBSD Core Team, TrustedBSD Projects robert@fledge.watson.org Senior Research Scientist, McAfee Research