From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Mon, 15 Oct 2007 18:47:30 +0300 From: Michael Shigorin To: sysadmins@lists.altlinux.org Message-ID: <20071015154730.GX13827@osdn.org.ua> Mail-Followup-To: sysadmins@lists.altlinux.org Mime-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Disposition: inline Content-Transfer-Encoding: 8bit User-Agent: Mutt/1.4.2.1i Subject: [Sysadmins] Fwd: [monit-dev] Re: Proposal: remove redirect support in http protocol test X-BeenThere: sysadmins@lists.altlinux.org X-Mailman-Version: 2.1.9 Precedence: list Reply-To: shigorin@gmail.com, ALT Linux sysadmin discuss List-Id: ALT Linux sysadmin discuss List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 15 Oct 2007 15:47:31 -0000 Archived-At: List-Archive: Здравствуйте. Коллеги, кто-нибудь ещё пользуется тестом HTTP в monit? Апстрим думает, стоит ли держать поддержку редиректов или выкинуть как излишнее (и несовершенное) усложнение. ----- Forwarded message from Martin Pala ----- Date: Thu, 11 Oct 2007 21:31:13 +0200 From: Martin Pala To: The monit developer list Subject: [monit-dev] Re: Proposal: remove redirect support in http protocol test Good point ... on the other side, there can be users already who use this test and thus will need to modify the configuration on upgrade. The redirect test in monit cvs can also identify the loops, which can be useful. So here is my +-0 (the proposed change has positive and negative features) Martin Jan-Henrik Haukeland wrote: >The HTTP protocol test supports rudimentary redirect within the same >server but not to external servers, though that should not be difficult >to add. However looking at the redirect code now I think the best would >be to remove it from monit. The reason is that it adds extra unnecessary >complexity without adding much in functionality and it also keeps the >socket open for keep-alive timeout. Users should instead write the >direct URL and we should send the "Connection: close" header as well >which would be easier on the server we test. For monit it does not >matter though since the socket is closed anyway after the http protocol >test returns. > >Jan-Henrik ----- End forwarded message ----- -- ---- WBR, Michael Shigorin ------ Linux.Kiev http://www.linux.kiev.ua/ ---- Oct 26--27, Kiev, Ukraine: -- http://conference.osdn.org.ua