I had a similiar problem with AD and it had to do with the base DN. I had set the base
DN at the root node and searches were slow and some failing altogether because
some nodes below my base were not accessible. Once I set the base to a level
just above where the users are stored, everything was OK.
The process hovers around 90% throughout this time and then once the page has been served dies back down.
The query logging is showing that Clearspace is doing a lot of group lookups. Even when the only thing I'm doing is watching the query logs page...
The LDAP settings I've used work fine with other apps (FishEye, JIRA, etc etc).
The AD can really slow down the site, so you need to be specific where where you set the base, as described in bdavis's post.
Sorry for the long response I was on vacation! I've set the base DN to cn=users,dc=domain,dc=com. I don't believe we have any subcontainers defined that we could point to. When looking at the Users section I see that there are right around 550 objects defined.
Would the recommendation be to separate out only the users into a sub container and point to that container?
Any suggestions would be most welcome!
I think I'm seeing your problem now. I enabled LDAP group support and the site pretty
much came to a halt. We have a lot of groups. I remember this never worked well with
Forums or Knowledge Base and I gave up on it. I think Clearspace is using the
LdapGroup manager by default. Here is another related thread:
Can you increase your user cache (to something like 5 Mb) to help with this issue? This may help.
I turned on debug and I can see that when I log in, Jive is looking up and creating
every single user in every group to which I belong. This amounts to several hundred
users who will never use Jive. Is this on purpose? If not, will this be fixed in a future
Thank you Bdavis! When I deleted that property I no longer had slow downs with my LDAP AD Integration. I consider this more of a work around rather then a full fix, but it gets me going!