This means that the same search, on the same data, can return different
results at different times, depending on whether or not a previous
search had succeeded and returned given attribute (samAccountName).
Currently, we have worked around this problem by either:
- running a bogus ldapsearch just after starting slapd, that returns
this attribute
- adding the attribute description to a local schema
What is the general opinion on working around this issue? Might there be
a way to implement something that ensures this happens automatically?
I'm thinking of delegating filter checking to proxied servers for ldap
backends, or automatically fetching schemas from them... Any experience,
ideas?