Rich Megginson wrote:
[Snip]
Ok, I see. I'm not sure if directory server auto-recovery is possible with this sort of failure. You might be able to use the berkeley db command line utilities to upgrade and/or recover your database files. http://www.oracle.com/technology/documentation/berkeley-db/db/utility/index....
- you may have to figure out which version of bdb you are using, then
find the appropriate documentation - rpm -qi db4 or rpm -qi db4-utils. You might try db2ldif to export your data, then re-import using ldif2db, but that would force a replication re-init anyway.
I have rebuilt the system and reinitialised the roots from the other multi-master. The directory server looks as though it is working properly and I can connect to the admin server from one of the other machines that is running the older version of the code (fedora-idm-console) but if I run the same command on the restored box then it will not do anything just sits there having typed in the command. How do I diagnose and fix this?
Howard.