Date: Wed, 13 Nov 2002 12:21:34 +1030 From: Greg 'groggy' Lehey <grog@FreeBSD.org> To: Robert Watson <rwatson@FreeBSD.org>, David O'Brien <obrien@FreeBSD.org> Cc: cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/contrib/lukemftpd - Imported sources Message-ID: <20021113015134.GQ46066@wantadilla.lemis.com> In-Reply-To: <Pine.NEB.3.96L.1021112121539.38530I-100000@fledge.watson.org> References: <20021112171203.GB59816@dragon.nuxi.com> <Pine.NEB.3.96L.1021112121539.38530I-100000@fledge.watson.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday, 12 November 2002 at 12:32:29 -0500, Robert Watson wrote: > > On Tue, 12 Nov 2002, David O'Brien wrote: > >> On Tue, Nov 12, 2002 at 09:15:53AM -0500, Robert Watson wrote: >>> I don't supposes there's any chance this addresses any of the serious >>> problems I identified on arch@ regarding the feature completeness and >>> documentation correctness of lukemftpd on FreeBSD? You seem not to be >>> responding to my follow-up e-mails asking what the status of the problems >>> is, and also seem not to have applied the documentation fixes submitted in >>> PRs. And you're clearly aware of at least on of the PR's, as you are the >>> owner of it. >> >> Robert what exactly do you want me to do? > > I want you to either do the work necessary to make lukemftpd a > reality in the base tree, or remove it. > >> Work with the vendor to fix things, or just pull everything off the >> vendor branch? Fine I'll do just that. > > I'm not really interested in talking about the exact means by which you > make lukemftpd real. If it means cleaning it up with the vendor before > enabling it in the tree, fine. If it means taking it off the vendor > branch, also fine. But if it's not ready to be in the tree, and you have > no plans to address the problems, before the release, then it shouldn't be > in the tree. Shouldn't all this have been obvious? David, what is it about this discussion that you have such difficulty understanding? >>> I am also very concerned regarding your changes to the warning message >>> I added to inetd.conf to suggest that the BSD ftpd be an alternate to >>> lukemftpd. >> >> Here is the change: >> RCS file: /home/ncvs/src/etc/inetd.conf,v >> diff -u -u -0 -r1.59 -r1.60 >> @@ -9,4 +9,4 @@ >> -# WARNING: lukemftpd does not support PAM, MAC, per-class nologin files, >> +# WARNING: lukemftpd does not utilize PAM, MAC, per-class nologin files, >> >> -# or any login.conf resource limits or features; use it only if this is >> +# or any login.conf resource limits or features. Use this ftpd only if this is >> >> -# appropriate for your environment. If you require these features, use >> +# appropriate for your environment. If you require these features, use >> >> -# the regular FreeBSD ftpd below. >> +# the alternate FreeBSD ftpd below. >> >> I felt the wording was a little harsh to LukeMftpd and thus on LukeM. I >> felt for developer-relations reasons it should be reworded a little. I >> also thought shorter sentances was better. > > I'm not making any attempt to disparage Luke or his work, and I'm > sensitive to the issue of appropriate phrasing. On the other hand, it's > not me who has introduced the daemon into the tree such that it requires > large warnings. Luke's a personal friend of mine, and at times I've found that the discussion has been a bit hard on him. I don't think this is the case here, though. It's fairly obvious that lukem hasn't done anything wrong by writing a daemon for NetBSD which doesn't include FreeBSD-only features. On the other hand, you're doing him a disservice by including it in the tree before it's ready. > As discussed on arch@, authenticating daemons supporting login mechanisms > in FreeBSD all provide the following services: > > (1) Support for PAM > (2) Support for login class resource limits > (3) Support for other login.conf features, such as per-user nologin files > (4) Support for OPIE > (5) Support for Kerberos (according to the README) > (6) Support for MAC > (7) Last, but not least importantly, documentation that refers to the > software by the correct name, rather than claiming its another piece > of software. > > Right now, lukemftpd does none of these. As such, I've told you I believe > lukemftpd is feature incomplete. If you aren't willing to do the work to > make it integrated into the supported login and account management > infrastructure, then you shouldn't have committed it to the tree. This should be true for all additions to the tree. > Look: I'm not saying lukemftpd is a bad piece of software. I'm just > asking that you either integrate it properly with FreeBSD, or stop > claiming that it's feature-ready. The least possible integration > you could do is to properly document it as not supporting standard > FreeBSD daemon features and to properly differentiate it from the > FreeBSD ftpd. In addition, you can't really implicitly claim it to be the primary ftpd until it's feature ready. On the whole, I think that it should be taken out for 5.0-RELEASE and reinstated only when it has the FreeBSD-specific features necessary to be the primary ftpd. Greg -- See complete headers for address and phone numbers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20021113015134.GQ46066>