I have a question that I cannot get a straight answer to, I realize everybody is new at this.. DSFW in a fake temp tree works great, but this is not a real world scenario (I have found that most TIDS and Documentation assume a fake isolated tree).

In a real world scenario, a DSFW server is going to be added into an existing tree and participate in eDirectory and replication (at least of the DSFW partition) etc. I am seeing the following behavior though I cannot find out why or pinpoint what is happening..

1) DSFW Server installed into production tree. Works Great. Have verified the full installation and seems to be running as planned.
2) Other OES2 servers handle LDAP authentication for multiple applications via LDAP. Services include FTP, a custom 3rd party app, etc. Also includes standard OES services that use LDAP like NSS, NCP, NAMCD, etc.

3) As long as the DSFW server is up and running, everything is fine.

4) If DSFW server is taken down for any reason, rebooted, or just shut off, ALL of the other services start to fail even though they are on a completely different server and referencing different LDAP sources (typically the local server they reside on).
A) NSS Volumes disappear and users are disconnected
B) FTP stops authenticating
C) Custom LDAP App fails
D) Anything else that relies on LDAP fails.

5) Bring DSFW server back up, services come back online and start working again (Some require restart but that's here nor there).

This is crazy because I can't find anything on these OES2 servers that have changed, all that happened is DSFW was introduced into tree. If DSFW server is pulled out (Recover from FAILED DSFW TID), the other services work normally as they should.

Why, when DSFW is introduced, do services fail and seemly start to rely on the DSFW server when they are not configured in any way to query the DSFW server? They are all set to query the other OES2 servers as they always did from the start.