Slashdot Mirror
Open Source X.500 Directory Projects?
DangerTenor asks: "The United States Government is standing up a Bridge Certificate Authority to enable PKI Interoperability between different agencies (gov't and non-gov't). The PKI currently relies on the use of either meta-directory products or X.500 DSP Chaining in order to pass certificates and CRLs between directories. OpenLDAP doesn't fit the bill because it doesn't support chaining. Does anyone know of open source projects focused on full X.500 directory implementation, or on meta-directory capabilities?"
A couple of search terms that you'd never come up with if you weren't already steep in the the arcane x.500 world: ISODE and quipu.
Most of this stuff comes out of and is maintained in Europe. As the RFC 1330 says,
"The ISODE is not proprietary, but it is not in the public
domain. This was necessary to include a "hold harmless"
clause in the release. The upshot of all this is that anyone
can get a copy of the release and do anything they want with
it, but no one takes any responsibility whatsoever for any
(mis)use."
You can still find the latest downloads via FUNET.
Be aware, this stuff is a major effort to compile and get working. It's big and complex, but well documented. Have fun, and let me know when you get dish -user "@c=$(COUNTRY)@o=ORG@cn=Manager" to give you a prompt.
Have you considered Novell's eDirectory it would do you proud, if your budget allows. It's not free but, it isn't expensive either.
-- Do you count?
Assuming that's what you're thinking about doing, I understand that Mozilla claims compatibility with OCSP, which is the bit that CAM uses proprietary software for. It also looks like generating signatures is also tied to the MS API in the CAM, but there's likely workarounds for that.
The OpenLDAP server actually comes with an LDAP backend. If it doesn't fit the bill, you could extend it. OpenLDAP is, after all, open source.
There's an old RFC that discusses this very thing: rfc2116
It's from '96, so it's probably incredibly out of date, but it might be a good place to start?
I would double check about OpenLDAP and chaining... I'm pretty sure it's at least on the development plan.
;-)
A much larger problem with OpenLDAP is scalability. OpenLDAP will not handle a large number of entries (+100k). OpenLDAP is a reference implementation of the LDAP RFCs and I don't think Kurt plans to complicate the implementation with what's required for scalability (connection pooling etc.).
The only usuable X.500 compatible directories other than OpenLDAP are all closed-source. Many are free though. I'd recommend taking a look at IBM Directory and Novell's eDirectory. There is much more involved in getting a directory environment going and having worked on Linux directories for IBM, I would of course recommend that out-source to experts to get things going
int func(int a);
func((b += 3, b));
AFAIK OpenLDAP is the only reasonably complete open source LDAP implementation. There have been many reports about OpenLDAP not scaling up to larger enterprises and missing features, and this is basically because of TANSTAAL.
Novell and iPlanet both sell working directory servers, but I don't know how well they support PKI, although I do know iPlanet supports SASL.
In any event, consider that there may not be a solution in this case. You are talking about a very specialized field with an audience which is corporate almost by definition.
well it might be worth a shot
d ir ectory/
have a look at
http://developer.apple.com/darwin/projects/open
I dont think its X.500 but they might have a plugin
regards
John Jones
The people that actually write the standards (like Dave Chadwick) appear to have given up on X500 as the trend is (VHS-like) towards LDAP. SO it's likely that LDAP will simply acquire the X500 attributes that made X500 so usable (but as always also too flexible to implement without having at least a clue about what you're doing). Give it time - or sponsor the guys that write the standards to do further work on it quicker ;-).
Insert