Dan,

Dan Melomedman <[EMAIL PROTECTED]> wrote:
> tinyldap is very, very young, and very limited since Fefe
> needs developers to extend it, and AFAIK has made very
> little progress on it since 2002, if any. He uses it,
> and it fits his needs, but it's not very useful to most
> anyone else. You must be able to update entries in an LDAP
> database to make it useful in a typical ISP environment.
> 
> If you are a coder, and like very clean code and design,
> then go ahead and help with tinyldap, email Fefe.

As I stated in my first post, I don't need to be able to update entries one
at a time -- I'll just rewrite the entire database every time I need it.
Ugly, but less ugly than OpenLDAP I think.

I e-mailed Felix and he seems to think that it will work. He says tinyldap
supports the basic features that are needed by qmail-ldap. And because
tinyldap uses a one-process per request architecture, it's not likely to
crash, burn and stop handling requests.

I've just asked him how likely something else might go wrong -- like a
particular piece of data causing tinyldap to crash, and thus that record is
not accessible.

(Important note: The CVS version is what I would use. He has done some
bugfixing and development since the latest tarball.)

I'd appreciate any input or experience that anyone has.

> If you aren't and want the best stability with OpenLDAP,
> then I would _strongly_ recommend the 1.3.x version
> (1.3.13, I believe) with the GDBM backend. Note however,
> that you will be limited to the now archaic version of the
> LDAP protocol - version 2. I've been running 1.3.13 for
> several years now with only a few database corruptions and
> small mishaps. For instance, right now the server isn't
> busy, and is answering requests, but all of its threads
> seem to be looping and eating 75% of the CPU time. This
> is how crappy the OpenLDAP code is!

Yuck!

> If you still want to use 2.x.x version, then I strongly
> recommend NOT using the Berkely DB backend. Use GDBM
> instead, you'll reduce the number of future database
> corruptions dramatically. You will also avoid the bulk
> of the bugs and Berkeley DB management overhead. If
> you've been hanging around the OpenLDAP mailing lists,
> you'll see many responses from OpenLDAP developers such
> as "This isn't our problem, this is Berkeley DB problem,
> go fix it instead".

I don't mind using GDBM or sticking with LDAP protocol v2.

Why do you recommend staying with 1.3.x? Was something broken when the 2.x.x
version was released? I believe you, but it makes me warry about OpenLDAP if
they are going backwards on stability.

Thanks,

David


Reply via email to