openldap-devel May 2008

openldap-devel@openldap.org

23 participants
23 discussions

connection_read_thread() return value
by Hallvard B Furuseth 16 May '08

16 May '08

slapd/connection.c:connection_read_thread() returns: - an int from connection_read() converted to long converted to void*, - or a void* (NULL in the places I found) converted to long to int to long to void*. Not sure what the idea is, but perhaps it should be cleaned up a bit:-) -- Hallvard

1 0

slapd/acl.c acl_set_cb_gather() strangeness
by Hallvard B Furuseth 16 May '08

16 May '08

Does slapd/acl.c:acl_set_cb_gather() make sense? It loops through the search attributes and sets bvalsp to each recognized attribute in turn without using it for anything else. After the loop it uses the final recognized attribute in slap_set_join(), thus ignoring previous attrs. -- Hallvard

2 1

[Fwd: Re: --enable-smbk5pwd support]
by Howard Chu 13 May '08

13 May '08

Comments on making contrib modules buildable from the main Makefile... -------- Original Message -------- Subject: Re: --enable-smbk5pwd support Date: Mon, 12 May 2008 08:03:49 +0200 From: manu(a)netbsd.org (Emmanuel Dreyfus) To: hyc(a)symas.com (Howard Chu), ando(a)sys-net.it (Pierangelo Masarati) Howard Chu <hyc(a)symas.com> wrote: > I have a couple reservations about this move. It's got some pretty major > external dependencies. Anything that's integrated into the built-in code > is expected to be fully autoconf'd, and this would require adding a lot of > tests for Kerberos etc. Also, we generally have to provide a test script > for any built-in components. And, we'd need to customize the enable switch > for it, or something, to toggle KRB5 or Samba support. All in all I hate > having to muck further with configure, and adding more tests to it that > just slow things down. Well, we can keep it a contrib thing, without full autoconf support, but help building it a bit. That will make packagers' life easier. Here is another proposal: add a --enable-contrib-overlays flag, which would add servers/slapd/overlays/contrib.c to the built files. That file would contain something like this: #ifdef CONTRIB_ADDPARTIAL #include "../../../contrib/slapd-modules/addpartial/addpartial.c" #endif #ifdef CONTRIB_ALLOP #include "../../../contrib/slapd-modules/allop/allop.c" #endif #ifdef CONTRIB_SMBK5PWD #include "../../../contrib/slapd-modules/smbk5pwd/smbk5pwd.c" #endif And so on... Packagers would just have one patch to that file, to do the right defines and includes required stuff so that it can build on their particular environement. A single patch to a simple file such as this one is much more easier to maintain than multiple patches on configure and Makefiles, which are fast moving targets. -- Emmanuel Dreyfus http://hcpnet.free.fr/pubz manu(a)netbsd.org -- -- Howard Chu CTO, Symas Corp. http://www.symas.com Director, Highland Sun http://highlandsun.com/hyc/ Chief Architect, OpenLDAP http://www.openldap.org/project/

6 8

slapo-nops
by manu＠netbsd.org 13 May '08

13 May '08

Hello I've written slapo-nops, an overlay that removes idempotent operations where attribute values are being replaced by the same values. This is a workaround to broken client that take the whole entry and change all attributes, even the ones that have not changed. Such behavior makes logs unreadable, and it badly weight on accesslog databases. Hence the overlay to filter the garbage out of modification operations. Is it desirable in servers/slapd/overlays/slapo-nops? In contrib/slapd-modules? Nowhere? -- Emmanuel Dreyfus http://hcpnet.free.fr/pubz manu(a)netbsd.org

5 5

RE_2_3 testing
by Quanah Gibson-Mount 12 May '08

12 May '08

Please test RE23 and let me know if you find any issues. This would hopefully be for the final 2.3 release. Thanks, Quanah -- Quanah Gibson-Mount Principal Software Engineer Zimbra, Inc -------------------- Zimbra :: the leader in open source messaging and collaboration

5 6

LDAPI and AutoBind
by Michael Ströder 11 May '08

11 May '08

HI! Lurking on the FDS list I noticed the new "Autobind" feature of FDS for LDAPI connections which directly emulates a SASL EXTERNAL bind if the client connects over LDAPI with a certain user-ID and simple bind (or no bind at all). It's configured at the server's side. See http://directory.fedoraproject.org/wiki/LDAPI_and_AutoBind Wouldn't that be a useful feature in OpenLDAP's slapd too for LDAP for automagically binding LDAP clients which aren't capable of sending SASL-Bind EXTERNAL but are capable to connect via LDAPI? Ciao, Michael.

3 2

Re: commit: ldap/servers/slapd config.c
by Howard Chu 09 May '08

09 May '08

hyc(a)OpenLDAP.org wrote: > Update of /repo/OpenLDAP/pkg/ldap/servers/slapd > > Modified Files: > config.c 1.488 -> 1.489 > > Log Message: > Use underscore in network_timeout I figured since this hadn't been documented yet I could change it without repercussions. But it raises a question of consistency; we have some options using hyphens, and some using underscores, and I don't really see any rationale for which is chosen. Anyone have a preference, and a reason? -- -- Howard Chu CTO, Symas Corp. http://www.symas.com Director, Highland Sun http://highlandsun.com/hyc/ Chief Architect, OpenLDAP http://www.openldap.org/project/

5 10

Re: commit: ldap/servers/slapd/overlays constraint.c
by Howard Chu 09 May '08

09 May '08

manu(a)OpenLDAP.org wrote: > Update of /repo/OpenLDAP/pkg/ldap/servers/slapd/overlays > > Modified Files: > constraint.c 1.14 -> 1.15 > > Log Message: > Add attribute size and count constraints to slapo-constaint > Approved by Ando > > Usage example: > overlay constraint > constraint_attribute jpegPhoto size 131072 > constraint_attribute userPassword count 3 Just as an aside, I'll note that back-ndb already requires maximum sizes to be configured for all attribute types. Currently it uses the {size} defined in the schema, and it also allows an explicit attrlen to be configured for specific attributes. It might be useful to write the size constraints configured here into the schema structure as well. -- -- Howard Chu CTO, Symas Corp. http://www.symas.com Director, Highland Sun http://highlandsun.com/hyc/ Chief Architect, OpenLDAP http://www.openldap.org/project/

2 1

Re: commit: ldap/servers/slapd config.c
by Quanah Gibson-Mount 09 May '08

09 May '08

--On Thursday, May 08, 2008 4:49 AM -0700 Howard Chu <hyc(a)symas.com> wrote: > hyc(a)OpenLDAP.org wrote: >> Update of /repo/OpenLDAP/pkg/ldap/servers/slapd >> >> Modified Files: >> config.c 1.488 -> 1.489 >> >> Log Message: >> Use underscore in network_timeout > > I figured since this hadn't been documented yet I could change it without > repercussions. But it raises a question of consistency; we have some > options using hyphens, and some using underscores, and I don't really see > any rationale for which is chosen. Anyone have a preference, and a reason? I strongly prefer hyphens. In part because they are treated as word boundaries in vi. ;) --Quanah -- Quanah Gibson-Mount Principal Software Engineer Zimbra, Inc -------------------- Zimbra :: the leader in open source messaging and collaboration

7 8

Re: commit: ldap/servers/slapd config.c
by Gavin Henry 08 May '08

08 May '08

>> The majority things use hyphens don't they? Please keep replies on list..heh ;-) > Dunno if there's a clear majority. There's plenty of both. > > Tho for the sake of keeping 2.4 self-consistent, we could revert this > patch > and fix the man pages. I don't really have any preference one way or the > other. Fair enough. If it's really worth deciding in the long run, we should start a new thread for a vote or something. Gavin. Suretec Systems is a limited company registered in Scotland. Registered number: SC258005. Registered office: 13 Whiteley Well Place, Inverurie, Aberdeenshire, AB51 4FP.

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

openldap-devel May 2008