DragonFly commits List (threaded) for 2005-09
Re: cvs commit: src/contrib/tcp_wrappers diag.c
On 9/16/05, Matthew Dillon <dillon@xxxxxxxxxxxxxxxxxxxx> wrote:
> :Modifying contrib/ code directly is not our policy, please use the
> :patching framework to achieve an alternative. Thanks!
> : -Hiten
> Well, except original (un-versioned) contrib code was never converted
> to the new policy. If the contrib directory is not versioned it probably
> doesn't have a README.DRAGONFLY in it either, is likely really ancient,
> and for expediency could be modified directly.
> Reading the later postings... before we go putting tcp-wrappers in base
> we should see how difficult it is to bring the latest vendor release
I forgot to mention the latest vendor release is "7.6-ipv6.4"
The "7.6" release without ipv6 support is ~8 years old.
Our contrib/tcp_wrapper began from vendor's "7.6" release and had
Kame's ipv6 changes, which should be quite stable.
FreeBSD contrib/tcp_wrapper/hosts_access.c rev1.3 commit log stated:
"...The base of patches are from KAME package and are actually daily
used for more than a year in several Japanese IPv6 environments. ..."
The question is, should we abandon Kame's ipv6 code and adopt vendor's
Since Kame's code is widely tested and is stable, IMHO, we should not
update to the latest vendor's release
> into our contrib framework.
> Matthew Dillon
Live Free or Die