Date: Thu, 26 Jun 2008 20:15:29 -0400 From: Wesley Shields <wxs@FreeBSD.org> To: Jo Rhett <jrhett@netconsonance.com> Cc: ports@FreeBSD.org, boris@tagnet.ru Subject: Re: split the quagga ports to remove unstable patches from quagga port Message-ID: <20080627001529.GK12581@atarininja.org> In-Reply-To: <A0DC84A9-10F1-4B8C-A9A1-AB17087A0373@netconsonance.com> References: <83EB55D5-11A0-44C9-A469-A5C5BD6D972C@netconsonance.com> <20080626155616.GI12581@atarininja.org> <A0DC84A9-10F1-4B8C-A9A1-AB17087A0373@netconsonance.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Jun 26, 2008 at 03:41:11PM -0700, Jo Rhett wrote: > > On Thu, Jun 26, 2008 at 08:38:00AM -0700, Jo Rhett wrote: > >> The current maintainer has indicated that he is holding up the quagga > >> 0.99.10 port until an unknown time when the TCP MD5 checksum patches > >> are again working in the tree. I don't think that this is the right > >> thing to do, as the TCP MD5 checksums are not necessary for the vast > >> majority of installations, and DO NOT work with a GENERIC kernel. > >> > >> I'd like to request that the port be split into two ports -- one with > >> a stock quagga installation, and the other with the MD5 checksum > >> patches, due to the instability and constant work on said patches > >> making the port unstable. (and being unavailable right now holds up > >> access to a major bugfix line) > > On Jun 26, 2008, at 8:56 AM, Wesley Shields wrote: > > You already have the option for TCPMD5 being on or off (off by > > default). The solution in my mind is not two separate ports, but > > what is used now: OPTIONS. If an option is broken why not mark it > > as such until it can be addressed? If the patches become a > > workable (yet still experimental) option the BROKEN line can be > > removed. The port can still keep using newer versions while the > > patches are being developed. > > > This makes perfect sense to me, but the Boris has said he won't do > that, and will not update the port until he has TCP MD5 working > properly. Given that response, I am requesting that we split the > ports so that he can focus on the TCP MD5 port. I'm adding Boris to the CC in case he has not seen this thread yet. I see no reason not to update this port. Using BROKEN in the case of TCPMD5 is the cleanest solution - it provides an updated port and the ability to work on the patches in the interim. Boris, can you please give this some consideration or explain the rationale behind not updating the port so we may better understand your perspective? Thanks! -- WXS
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080627001529.GK12581>