There is an old draft for: https://tools.ietf.org/html/draft-ietf-boreham-numsubordinates-01
And I do think this type of attribute does supply value. Many client we work with monitor container counts for stats and compliance. Since I know Michael loves doing Drafts, perhaps he would like to ... ᐧ -- -jim Jim Willeke On Thu, Dec 11, 2014 at 4:42 AM, Emmanuel Lécharny <[email protected]> wrote: > Le 11/12/14 09:34, Michael Ströder a écrit : > > Emmanuel Lécharny wrote: > >> Le 11/12/14 05:32, amit nanda a écrit : > >>> I would like to get the total count of entries under a container. > >>> I application will have around 100000 entries, i can't get a direct > way to > >>> to this. > >> Plain normal. > > In web2ldap I'm using various mechanisms for e.g. getting the number of > > entries in a sub-tree. > > > > No-op search control > > > > Search and sum up server-specific operational attributes like: > > - subordinateCount > > - numSubordinates > > - numAllSubordinates > > - msDS-Approx-Immed-Subordinates > > That's really very server dependent. > > > > >> "In the control returned to the client, the size MAY be set to the > >> server's estimate of the total number of entries in the entire result > >> set. Servers that cannot provide such an estimate MAY set this size to > >> zero (0)." > >> > >> (RFC 2696). > > Interesting. I will check whether MS AD returns it. > > I'm not sure OpenLDAP returns it. We don't. > > > >> Side remark : it would be quite interesting to be able to provide such a > >> *real* number of entries that would be returned, without actually > >> returning any of those entries. The work would be done on the server, > >> and returned as a control. > > How about also implementing No-op search control? > > > Yes, that's an idea. Do we have any spec for that control ? > > > > In web2ldap I'm using it to display number of entries returned by a > search, > > number of entries affected by bulk modify/delete operations. I've > extensively > > tested it with OpenLDAP's slapo-noopsrch. > > > > However there's no I-D yet. > Ah, too bad... > > Thanks > >
