In the Directory Server 9.0 Release Notes in 2.2 is stated:
By moving the changelog database into the main database environment, the overall Directory Server performance is improved.
However, looking at a vanilla installation of Directory Server 9 on one of our systems and enabling the changelog with default settings, I still see the changelog existing in a separate directory.
[root@hydrogen slapd-ul]# pwd /var/lib/dirsrv/slapd-ul [root@hydrogen slapd-ul]# ls -la total 24 drwxrwx---. 6 rhds rhds 4096 Nov 29 09:20 . drwxr-xr-x. 4 root root 4096 Nov 29 09:11 .. drwxrwx---. 2 rhds rhds 4096 Nov 29 09:11 bak drwxr-xr-x. 2 rhds rhds 4096 Feb 8 15:54 changelogdb drwxrwx---. 4 rhds rhds 4096 Feb 10 05:35 db drwxrwx---. 2 rhds rhds 4096 Jan 10 14:25 ldif [root@hydrogen slapd-ul]#
As you can see, the main db files are in the db dir while the changelog still seems to exist in changelogdb.
In what manner is the merging of the changelog into the main database environment exactly arranged?
The reason I ask is because for one of our instances in Directory Server 8.2 we have the changelog on a separate filesystem to improve performance, and I wonder if this will still be necessary when migrating to 9.
Do you believe that there's still a performance gain to be made with putting the changelog db files on a separate filesystem on a separate physical disk than the main db files?
Red Hat Directory Server (RHDS) 9.0
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.