If we're idle, there might be objects pending cleanup for the last two
epochs. Unless another thread comes in and checks into a new epoch or we
shut down, they will linger forever.
If one of the objects was a connection, it wouldn't get closed and be
stuck in CLOSE_WAIT state, potentially refusing another ligitimate
connection if its socket address were to match the one we're yet to
close.
Reject operations in such a case with LDAP_BUSY. If read_event feature
is on, just stop reading from the connection. However this could still
result in deadlocks in reasonable situations. Need to figure out better
ways to make it safe and still protect ourselves.
Similar to the algorithm presented in
https://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-579.pdf
Not completely lock-free at the moment. Also the problems with epoch
based memory reclamation are still present - a thread actively observing
an epoch getting stuck will prevent LloadConnections and LloadOperations
being freed, potentially running out of memory.
LDAP_UNAVAILABLE signals "the server is shutting down or a subsystem
necessary to complete the operation is offline", so intelligent clients
tend to infer the connection will not be usable any more, which is not
the case here.
Unregistration is a hack and we shoould either make the subsystems into
an entry (if monitor allows subentry generation) or implement subsystem
unregistration in back-monitor.
An operation is rejected iff it has to be dropped before we can find an
upstream for it (unless we handle it ourselves, that is). At that point
it is failed unless completed successfully.
This makes a difference for multi-stage binds which alternate between
'failed' (we are waiting on a server response) and 'completed' (server
did what we asked them to, waiting on client to continue).
