Mailing-List: contact dev-help@httpd.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@httpd.apache.org
Message-ID: <53275716.60701@apache.org>
Date: Mon, 17 Mar 2014 21:12:06 +0100
From: Ruediger Pluem <rpluem@apache.org>
User-Agent: Mozilla/5.0 (X11; Linux i686 on x86_64;
 rv:27.0) Gecko/20100101 Firefox/27.0 SeaMonkey/2.24
MIME-Version: 1.0
To: dev@httpd.apache.org
Subject: Re: [VOTE] Release httpd 2.2.27 as GA?
References: <20140313155547.4868f940@hub>
	<CALK=YjOh6kuSk61CR0PYbrtk4jboxH2mG8-8m=pA6Asn7DTLcg@mail.gmail.com>
	<AB1691BE05AE7F4992697F2A0835627AA6BF9932@VOEXM10W.internal.vodafone.com>
	<5326E9BC.7050208@kippdata.de> <20140317141556.59529fd1@hub>
 <532751AE.2050305@kippdata.de>
In-Reply-To: <532751AE.2050305@kippdata.de>
Content-Type: text/plain; charset=ISO-8859-15
Content-Transfer-Encoding: 8bit


Rainer Jung wrote:
> On 17.03.2014 20:15, William A. Rowe Jr. wrote:
>> On Mon, 17 Mar 2014 13:25:32 +0100
>> Rainer Jung <rainer.jung@kippdata.de> wrote:
>>
>>> On 17.03.2014 10:59, Pl�m, R�diger, Vodafone Group wrote:
>>>> I try to find a slot to vote on 2.2.x later today, but I guess it
>>>> will be in the evening my local time (GMT+1).
>>>
>>> Same here, some builds and tests still running, currently looks OK.
>>>
>>> One minor nag: out of tree build found another problem in addition to
>>> the known one for bundled APR 1.5.0. When building httpd 2.2.27, the
>>> bundled libtool has version 1.5.26 which does not create the "tools"
>>> directory before trying to write tools/.libs/gen_test_char.o
>>>
>>> httpd 2.4.9 contains libtool 2.4.2, which does not exhibit that
>>> problem.
>>>
>>> So we should also add
>>>
>>> 	$(APR_MKDIR) tools
>>>
>>> to the apr Makefile to avoid that pitfall. I'll fix that later in apr
>>> svn. Basically the breakage of out of tree build for APR 1.5.0 was
>>> known (r1541744), this is just a new detail.
>>
>> Agreed that's the fix, and there has been chatter about putting out
>> 1.5.1 one of these days.  I don't think this is enough to warrant
>> dumping the release, but I was afraid something odd might happen by
>> picking up 1.5.0 over 1.4.8.  That said - I think 1.5.0 was still the
>> right call for the time being.
>>
>> Anyone doing out of tree builds aught to be able to read the output
>> of make >build.log 2>&1, and work that out.  And most who build out
>> of tree I expect also build apr, apr-util individually.
>>
>> The casual user who would be bit by this bug is undoubtedly grabbing
>> the tarball and just building the entire stack in-tree, so I'm not
>> worried to much.
> 
> +1 to that.

+1

Regards

R�diger