From owner-freebsd-ports@FreeBSD.ORG Mon Apr 9 23:01:59 2012 Return-Path: Delivered-To: freebsd-ports@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id CE155106566C for ; Mon, 9 Apr 2012 23:01:59 +0000 (UTC) (envelope-from freebsd-ports@herveybayaustralia.com.au) Received: from mail.unitedinsong.com.au (mail.unitedinsong.com.au [150.101.178.33]) by mx1.freebsd.org (Postfix) with ESMTP id 46CA18FC12 for ; Mon, 9 Apr 2012 23:01:59 +0000 (UTC) Received: from laptop1.herveybayaustralia.com.au (laptop1.herveybayaustralia.com.au [192.168.0.182]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) by mail.unitedinsong.com.au (Postfix) with ESMTPSA id 208D75C22 for ; Tue, 10 Apr 2012 09:15:35 +1000 (EST) Message-ID: <4F836A65.7080402@herveybayaustralia.com.au> Date: Tue, 10 Apr 2012 09:01:57 +1000 From: Da Rock User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:7.0.1) Gecko/20111109 Thunderbird/7.0.1 MIME-Version: 1.0 To: freebsd-ports@freebsd.org References: <4F7E498E.7070007@herveybayaustralia.com.au> <4F804679.2040803@herveybayaustralia.com.au> <4F80D4F9.9020207@herveybayaustralia.com.au> In-Reply-To: <4F80D4F9.9020207@herveybayaustralia.com.au> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: FreeBSD Port: samba34-3.4.14 X-BeenThere: freebsd-ports@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting software to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 09 Apr 2012 23:02:00 -0000 On 04/08/12 09:59, Da Rock wrote: > On 04/08/12 00:02, Chris Rees wrote: >> On 7 April 2012 13:51, Da >> Rock wrote: >>> On 04/06/12 13:20, Timur I. Bakeyev wrote: >>>> Hi! >>>> >>>> Can you show the build output? Port should't directly dipend from >>>> anything like this. >>> Sorry, it was on a clients system and I couldn't get the info >>> across. I can >>> only say for sure that installing libnet fixed the issue. >>> >>> The build error was an undeclared function or const (most likely >>> const) that >>> was supplied by libnet. >>> >>> Sorry I can't tell you any more information; I emailed this to >>> notify of the >>> fix, and for others if they come across the issue in the future. >>> Googling >>> was what put me on to this fix as well. >> If you could give some idea of the site that advised this, that could >> probably help too. > > Again, my apologies. Let me try this again and see if I can be a > little more organised with my description. > > This was a clean, fresh install on an amd64 (Atom cpu, although I did > try on a real cpu too). I tried 3.6 first, it didn't build. Then I > tried 3.5, when that didn't work (same error) I looked at the handbook > and it was still on 3.4, and during my googling I noticed that there > could be an issue with incompatibility with the libsmbclient; so I > reverted to 3.4. It didn't build either, and they all had the same error. > > So I googled some more on this error, but there wasn't much on it at > all. I did notice a lot of comments on samba, tdb, and libnet, and > there was a clue in the error of a libnet dependency (sorry I just > can't remember, so much has happened in the 24 hr period); so I tried > installing libnet and then built samba 3.4 (to avoid further issues > and conflicts) and presto! it all built. > > That was the day prior to my post, and I figured someone could run a > clean build and find this as well; failing that, someone would come > across this again in the future. As you can see it wasn't anything in > particular that set me on this, just a hunch I followed and lucked out > on a resolution, but it worked :) > > I hope that helps someone... To drag this up again, I was thinking about the number of cases I've found like this recently, and I was considering what the most appropriate action to take here. This one is obviously controversial, and I didn't have the time to do more or test further, but for future reference I'd like some clarification. I'd say a PR is not really appropriate as a response to an issue such as this (unless the maintainer offers no response at all), but should I create a patch to assist the maintainer? Or is that over doing it? If I were to create a patch, what is the correct (usable) procedure? And for something like this it would be an adjustment to BUILD_DEPENDS, correct? Thanks for the clarification guys.