[389-users] Recovery Strategy

[Gerrard Geldenhuis]

Hi
As far as I can see the documentation does not make mention of backups other than the userdb, netscapedb and dse.ldif.

With regards to the certificate databases and admin server configuration is there any specific strategies, recommendations or readmade scripts?

I am looking at scenarios where we would lose a server completely. I have considered two possible ways of recovering and would appreciate any thoughts ,recommendations or warnings of peril.

Backup the following files:
tar :
/etc/dirsrv/slapd-<instance>/dse.ldif
/etc/dirsrv/slapd-<instance>/pin.txt
/etc/dirsrv/slapd-<instance>/*.db
/etc/dirsrv/admin-serv/*
/var/lib/dirsrv/slapd-ie1auth002/bak/*

Recovery method #1:
yum install 389-ds -y
untar all files and create directories.
use bak2db to restore databases
service dirsrv start

Recovery method #2:
Summary: Install software and "build from scratch"

Gory detail:
yum install 389-ds -y
setup-ds.pl -f settings.file -s
Copy *.db files back
enable-changelog
create netscape root suffix
enable master replica for netscape root
re-initiate from other master
register-ds-admin -f setingsfile -s -u
There would be two ways of getting data back either be re-populating from other databases or by restoring the backed up data using bak2db.

I have not yet tested these methods so there might be omissions in the steps above.

Best Regards




________________________________________________________________________
In order to protect our email recipients, Betfair Group use SkyScan from 
MessageLabs to scan all Incoming and Outgoing mail for viruses.

________________________________________________________________________
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.fedoraproject.org/pipermail/389-users/attachments/20100920/bbb9ed5a/attachment.html>