Return-Path: Delivered-To: apmail-activemq-users-archive@www.apache.org Received: (qmail 15212 invoked from network); 29 Jan 2009 17:58:02 -0000 Received: from hermes.apache.org (HELO mail.apache.org) (140.211.11.2) by minotaur.apache.org with SMTP; 29 Jan 2009 17:58:02 -0000 Received: (qmail 65258 invoked by uid 500); 29 Jan 2009 17:58:01 -0000 Delivered-To: apmail-activemq-users-archive@activemq.apache.org Received: (qmail 65241 invoked by uid 500); 29 Jan 2009 17:58:00 -0000 Mailing-List: contact users-help@activemq.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: users@activemq.apache.org Delivered-To: mailing list users@activemq.apache.org Received: (qmail 65227 invoked by uid 99); 29 Jan 2009 17:58:00 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Thu, 29 Jan 2009 09:58:00 -0800 X-ASF-Spam-Status: No, hits=0.2 required=10.0 tests=SPF_PASS,WHOIS_MYPRIVREG X-Spam-Check-By: apache.org Received-SPF: pass (athena.apache.org: domain of rajdavies@gmail.com designates 74.125.78.145 as permitted sender) Received: from [74.125.78.145] (HELO ey-out-1920.google.com) (74.125.78.145) by apache.org (qpsmtpd/0.29) with ESMTP; Thu, 29 Jan 2009 17:57:51 +0000 Received: by ey-out-1920.google.com with SMTP id 4so10163eyg.26 for ; Thu, 29 Jan 2009 09:57:30 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:from:to :in-reply-to:content-type:content-transfer-encoding:mime-version :subject:date:references:x-mailer; bh=G1b1gGdx75ePmb0khtzZBEhF//APkg82tKum55C8iw4=; b=sx1uPKW86A3NV/ls1ucGGt1TV1iVYzFLihhBMMM/72HbpNKYSUF090rs6Q/6ZhzRZP /r64ghSF4Wyb9Ib7oTNkZqzStrqZtQ5L/S78LI4dRPXHuZqded5tIrr0ti+DsArhi4m7 RTmRwpu4P7abakdxTUQzYBPjkhwp+UdGkke5o= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:from:to:in-reply-to:content-type :content-transfer-encoding:mime-version:subject:date:references :x-mailer; b=sn8lEQrliZQ8PxHXzv94/tRN4tBMzH/hMIdfg7wNmqGtnVfzNGZIse3NuuaIXHjBB2 F7+Ts/+CdC09VyNEwu/aqWqHoSslwj6IHE6KKn/1c8Ct8gCoTmM+MfUVKmZitPusjLQG 37Jz2ltvdgwWAi8bn8bTLT52VrWFFUjUESj9w= Received: by 10.210.42.13 with SMTP id p13mr312550ebp.183.1233250211945; Thu, 29 Jan 2009 09:30:11 -0800 (PST) Received: from ?192.168.1.76? (host86-168-83-211.range86-168.btcentralplus.com [86.168.83.211]) by mx.google.com with ESMTPS id 31sm30308nfu.22.2009.01.29.09.30.08 (version=TLSv1/SSLv3 cipher=RC4-MD5); Thu, 29 Jan 2009 09:30:09 -0800 (PST) Message-Id: <44508F78-A457-4A64-A915-82683C829E22@gmail.com> From: Rob Davies To: users@activemq.apache.org In-Reply-To: Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v930.3) Subject: Re: Single consumer for multiple brokers? Date: Thu, 29 Jan 2009 17:30:05 +0000 References: <15768836.post@talk.nabble.com> <15768993.post@talk.nabble.com> <15769001.post@talk.nabble.com> <15769003.post@talk.nabble.com> <15769009.post@talk.nabble.com> <21729358.post@talk.nabble.com> <3a73c17c0901290813j59f5dd20k2a665df61c42a0ee@mail.gmail.com> X-Mailer: Apple Mail (2.930.3) X-Virus-Checked: Checked by ClamAV on apache.org http://activemq.apache.org/fanout-transport-reference.html - is the place to look to get started On 29 Jan 2009, at 17:00, James Strachan wrote: > The fanout transport kinda has most of the implementation code to do > this; it just needs to be hacked a little so that consume start/stop > commands are sent to multiple brokers and transactions/acks are > correctly sent to the broker that sent the original message. There > might be some fun and games with transactions in there mind; but for > simple-ish use cases it shouldn't be too hard to hack the > FanoutTransport if anyone fancies having a go > > > 2009/1/29 Jim Lloyd : >> I'm interested in a variation of this for load balancing. Assume >> the volume >> of data coming from publishers is very large, and you need N >> brokers (think >> N is 10 or more). Furthermore, you want N+1 or N+2 redundancy so >> that if any >> 1 or 2 brokers die, there is sufficient capacity to continue >> handling all of >> the traffic (after affected publishers reconnect). The brokers >> might all be >> behind a hardware load balancer (e.g. F5 or NetScaler) so that all >> publishers connect via a virtual IP. >> >> Now, from the consumer side, we also have many consumers. Each >> consumer only >> subscribes to a subset of the available topics, and we arrange via >> design of >> our topics such that one consumer can always keep up with the >> volume of data >> published on one topic. But the consumer must connect to every >> broker. >> >> So, we need a fan-in variation of a failover transport that >> connects to >> every broker and actively consumes data from all brokers. If any >> broker >> disconnects, the consumer would use the exponential backoff >> reconnect logic >> to reconnect when the broker becomes available. >> >> So, the backup=true option isn't helpful. Instead, we want >> something like >> "fan_in=true". >> >> It's not hard to build this kind of fan in logic on top of the >> failover >> transport, but it would be cool if the failover transport was >> capable of >> doing fan in directly. >> >> On Thu, Jan 29, 2009 at 8:13 AM, Gary Tully >> wrote: >> >>> have you looked at the failover transport? see: >>> http://activemq.apache.org/failover-transport-reference.html >>> with the backup=true option, a connection to all listed brokers will >>> be created so that they are in hot standby in >>> the event that the first connection is dropped. In this way, >>> failover >>> can be very fast. >>> >>> 2009/1/29 kaykay : >>>> >>>> This thread is old but just curious if there has been a recent >>>> update of >>> this >>>> w.r.t ActiveMQ 5.2 . The problem that I am trying to solve is >>>> similar >>> where >>>> a consumer listens to multiple brokers (as a failover redundancy >>>> issue >>>> instead of listening to a single broker). >>>> >>>> >>>> Stepan Koltsov wrote: >>>>> >>>>> I'm playing with 5.0. >>>>> >>>>> How do you think, is it hard to write new Transport that >>>>> consumes from >>>>> multiple brokers? >>>>> >>>>> S. >>>>> >>>>> >>>>> ttmdev wrote: >>>>>> >>>>>> Yes, I got similar results in my consumer testing. What version >>>>>> of AMQ >>>>>> are you using? >>>>>> >>>>>> Your only recourse may be to multi thread your consumer and >>>>>> have it >>>>>> create a connection to each of the brokers. >>>>>> >>>>> >>>>> >>>> >>>> -- >>>> View this message in context: >>> http://www.nabble.com/Single-consumer-for-multiple-brokers--tp15768836p21729358.html >>>> Sent from the ActiveMQ - User mailing list archive at Nabble.com. >>>> >>>> >>> >>> >>> >>> -- >>> http://blog.garytully.com >>> >>> Open Source SOA >>> http://FUSESource.com >>> >> > > > > -- > James > ------- > http://macstrac.blogspot.com/ > > Open Source Integration > http://fusesource.com/