Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 12 Aug 1998 23:43:10 +0000 (GMT)
From:      Terry Lambert <tlambert@primenet.com>
To:        mike@smith.net.au (Mike Smith)
Cc:        pvh@leftside.wcape.school.za, freebsd-hackers@FreeBSD.ORG
Subject:   Re: Moving ahead with LDAP
Message-ID:  <199808122343.QAA13385@usr05.primenet.com>
In-Reply-To: <199808122113.OAA00425@dingo.cdrom.com> from "Mike Smith" at Aug 12, 98 02:13:38 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> Are there any portable LDAPv3 servers available?  Is the UMich server 
> at a developmental dead-end?

The list was very dead for a long time, until I kicked it back in
March or thereabouts.

After that, a number of Linux people beat on me after getting my
patches until I looked at their problems and fixed them.

Basically, before I came along, the code was in very poor shape
and had some bad assumptions about threading and select(2)'s
first argument, etc..  After I took all of the Critical Angle
patches and integrated them, fixed the dbm backend namespace
bug, and fixed the threads and select assumptions, the code
started working on about 8 platforms (SGI uses Draft 4 pthreads
too).

Since then, the Linux guys have gone nuts... there is a Linux
specific LDAP WWW site dedicated to doing an LDAP-based Linux
parameter store.  Ah, to be a student and get back half my
day again while scholarships and grants paid my way... ;-).

Anyway, it appears that many of the UMICH people have gone on to
commercial pastures, where they have a bit of a vested interest in
keeping the patches hard to apply and the protocol one or more
revisions out of step.

I don't hold out much hope for the Linux effort; the people involved
are enthusiastic, but were posting such things as "how do I fix
this ``sys_errlist'' compile problem?" and generally stumbling
over gdbm not woking, but the Berkely dbm working, and making
the same errors about threading models that the original authors
made (threading is considered preemptive if a blocking select() will
result in a context switch, but they were tagging pthreads as
being non-preemptive, and buzz-looping as a result).

I'd like to see a serious effort at a commercially usable v3
server implementation (hopefully carrying forward under the UMich
license or a UCB style project license for independently developed
pieces and the agregation of the new and the old).

I'm not sure the boolean.net repository promotes this; mostly
because I'm not that sure that the UMich code is capable of evolving
into a v3 server.


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199808122343.QAA13385>