Re: Major new ubiqx release

Luke Kenneth Casson Leighton (lkcl@regent.push.net)
Wed, 22 Apr 1998 12:37:00 +0000 (GMT)

Date:	Wed, 22 Apr 1998 12:37:00 +0000 (GMT)
From:	Luke Kenneth Casson Leighton <lkcl@regent.push.net>
To:	"Christopher R. Hertel" <crh@NTS.Umn.EDU>
Subject: Re: Major new ubiqx release
In-Reply-To: <199804212156.QAA12241@unet.unet.umn.edu>

On Tue, 21 Apr 1998, Christopher R. Hertel wrote:

> You misunderstand...
>
> Let's say, for example, that someone writes an interface that copies all
> of the WINS entries into an LDAP server database. Now, someone with
> authority comes along and modifies an entry via LDAP. Do we allow that
> modification to be sent back to nmbd or not?

hm. there can only be read-cacheing of the WINS entries in the LDAP
database for practical reasons. so it's not so much a matter of "sent
back" in this particular instance.

direct answer: yes.

hm. i suppose i was kind of thinking of a "proprietary" database system,
where it didn't have some form of replication, such that
read-and-write-cacheing could be done (e.g the wins.dat mechanism).

good point, chris.