Re: [discussion] Release 2.0.65 [the final frontier]
Hi Bill, On 02.07.2013 01:47, wr...@rowe-clan.net wrote: I am not at all concerned whether APR 0.9 is released again or not since folks had years to take that up in our discussions of putting httpd 2.0 to bed, yet nobody so much as suggested a release, nevermind some volunteer to act on it. true; but I thought that most of us probably forgot about that we bundle APR/APU with 2.0.x - like I did; the lack of APR/APU fixes came only to my attention when I was on building the 2.0.65 binaries ... but since nobody else expressed an oppinion about then thats fine, and I shut up. or if you have concurred with the group consensus to let this story end as of Jun 2013. I have. Just did put the NetWare bins up; go ahead and release. Gün.
Re: [discussion] Release 2.0.65 [the final frontier]
Hi, Maybe the simple option is to do the final release with the old/existing bundled APR, but put a foot note in the release notes that the newer APR v1.4.8/1.5.2 has been confirmed to successfully work with 2.0.65. This way it may give confidence to anyone who is stuck on 2.0.x for some reason to use the newer APR/APR-util if needs be. Regards, Mike On 02/07/2013 13:06, Guenter Knauf wrote: Hi Bill, On 02.07.2013 01:47, wr...@rowe-clan.net wrote: I am not at all concerned whether APR 0.9 is released again or not since folks had years to take that up in our discussions of putting httpd 2.0 to bed, yet nobody so much as suggested a release, nevermind some volunteer to act on it. true; but I thought that most of us probably forgot about that we bundle APR/APU with 2.0.x - like I did; the lack of APR/APU fixes came only to my attention when I was on building the 2.0.65 binaries ... but since nobody else expressed an oppinion about then thats fine, and I shut up. or if you have concurred with the group consensus to let this story end as of Jun 2013. I have. Just did put the NetWare bins up; go ahead and release. Gün.
Re: [discussion] Release 2.0.65 [the final frontier]
On Tue, Jul 2, 2013 at 8:53 AM, MikeM michaelm12-asfbugzi...@aquaorange.net wrote: Hi, Maybe the simple option is to do the final release with the old/existing bundled APR, but put a foot note in the release notes that the newer APR v1.4.8/1.5.2 has been confirmed to successfully work with 2.0.65. This way it may give confidence to anyone who is stuck on 2.0.x for some reason to use the newer APR/APR-util if needs be. APR/APR-util 1.x won't work with httpd 2.0.x. Someone continuing to use 2.0.x will need to hand-pick or backport fixes from apr/apr-util 0.9.x or later levels. But then they'll have to backport fixes from httpd too. The line was drawn at slightly different places for httpd vs. apr/apr-util, but the long term picture is the same: There is effort to remain on httpd 2.0.x if you want to pick up any code fixes, and the recommendation is clear. Regards, Mike On 02/07/2013 13:06, Guenter Knauf wrote: Hi Bill, On 02.07.2013 01:47, wr...@rowe-clan.net wrote: I am not at all concerned whether APR 0.9 is released again or not since folks had years to take that up in our discussions of putting httpd 2.0 to bed, yet nobody so much as suggested a release, nevermind some volunteer to act on it. true; but I thought that most of us probably forgot about that we bundle APR/APU with 2.0.x - like I did; the lack of APR/APU fixes came only to my attention when I was on building the 2.0.65 binaries ... but since nobody else expressed an oppinion about then thats fine, and I shut up. or if you have concurred with the group consensus to let this story end as of Jun 2013. I have. Just did put the NetWare bins up; go ahead and release. Gün. -- Born in Roswell... married an alien... http://emptyhammock.com/
Re: [discussion] Release 2.0.65 [the final frontier]
Hi Oh I see - I had not realised this. In that case, I agree that sticking with 0.9.x is the only sensible option at this point in time :) Mike On 02/07/2013 14:35, Jeff Trawick wrote: On Tue, Jul 2, 2013 at 8:53 AM, MikeM michaelm12-asfbugzi...@aquaorange.net mailto:michaelm12-asfbugzi...@aquaorange.net wrote: Hi, Maybe the simple option is to do the final release with the old/existing bundled APR, but put a foot note in the release notes that the newer APR v1.4.8/1.5.2 has been confirmed to successfully work with 2.0.65. This way it may give confidence to anyone who is stuck on 2.0.x for some reason to use the newer APR/APR-util if needs be. APR/APR-util 1.x won't work with httpd 2.0.x. Someone continuing to use 2.0.x will need to hand-pick or backport fixes from apr/apr-util 0.9.x or later levels. But then they'll have to backport fixes from httpd too. The line was drawn at slightly different places for httpd vs. apr/apr-util, but the long term picture is the same: There is effort to remain on httpd 2.0.x if you want to pick up any code fixes, and the recommendation is clear. Regards, Mike On 02/07/2013 13:06, Guenter Knauf wrote: Hi Bill, On 02.07.2013 01:47, wr...@rowe-clan.net mailto:wr...@rowe-clan.net wrote: I am not at all concerned whether APR 0.9 is released again or not since folks had years to take that up in our discussions of putting httpd 2.0 to bed, yet nobody so much as suggested a release, nevermind some volunteer to act on it. true; but I thought that most of us probably forgot about that we bundle APR/APU with 2.0.x - like I did; the lack of APR/APU fixes came only to my attention when I was on building the 2.0.65 binaries ... but since nobody else expressed an oppinion about then thats fine, and I shut up. or if you have concurred with the group consensus to let this story end as of Jun 2013. I have. Just did put the NetWare bins up; go ahead and release. Gün. -- Born in Roswell... married an alien... http://emptyhammock.com/
Re: [discussion] Release 2.0.65 [the final frontier]
On 30 Jun 2013, Guenter Knauf observed; it seems a bit odd to me that we now roll the 2.0.65 final without having APR/APU picking up latest fixes [1][2], making this release hanging around for ever bundled with APR/APU 0.9.x versions which lack latest stuff: And in six months, some other thing will have changed, and whatever code is sitting around 'forever' will be equally stale. I am not at all concerned whether APR 0.9 is released again or not since folks had years to take that up in our discussions of putting httpd 2.0 to bed, yet nobody so much as suggested a release, nevermind some volunteer to act on it. APR 0.9 is effectively retired already, and APR 1.0 was released how long ago? If someone at the APR project wanted to tag and roll 0.9.x (even designating it 'final' if that was the group's desire) that would be something for that project to work out). So far as I'm concerned, this httpd release is already long out-of-date, and is simply a polite way of offering one last bit of wiggle room for users who still haven't, and still will need to migrate to 2.2 as soon as reasonably possible, unless their httpd instances aren't actually accessible through untrusted networks or by untrusted users, such as the internet. Whether it fixes every last issue of any particular category doesn't worry me. The package 2.0.65 is tagged and our packages are never 'revisioned' after the fact, so there is no way to change the embedded versions. That is the blessing of dropping embedded apr libraries with the 2.4.x family of releases. So without dumping and rerolling and testing another release, we can't really accomplish what you suggest. As far as a release vote for apr 0.9, you would want to take that up with dev@apr. As far as this package goes, I count 3 +1's and will add my +1 when I announce the results, but 1) I'm on vacation and 2) it's now the 2nd half of 2013, years after we replaced 2.0 with 2.2, so I'm unwilling to do anything further with the 2.0.x source branch (including test another release candidate), other than helping wordsmith our end-of-2.0 consensus communications, and leave this image hanging on my office wall for posterity... http://www.gocomics.com/foxtrot/2005/01/11/ Please let us know whether you are personally taking on some 2.0.66 release project, or if you have concurred with the group consensus to let this story end as of Jun 2013. I'd like to avoid yet-another-vote over something the group has clearly and repeatedly expressed consensus over.
