upgrade from slapd from dapper to hardy fails on DB conversion
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
openldap2.3 (Ubuntu) |
Invalid
|
Undecided
|
Unassigned |
Bug Description
I just tried to upgrade a server with openldap from dapper to hardy (I pulled an image from the dapper server and am doing a dry-run in a virtualbox now).
During the upgrade, the slapd postinst script somehow crashed in an intermediate state, and the upgrade aborted shortly after that.
When I try "dpkg --configure slapd" now, I get this:
root@newsauron:
Setting up slapd (2.4.7-6ubuntu3) ...
Backing up /etc/ldap/
Upgrading BDB 'checkpoint' options... .
Moving old database directories to /var/backups:
Backup path /var/backups/
dpkg: error processing slapd (--configure):
subprocess post-installation script returned error exit status 1
Errors were encountered while processing:
slapd
After removing the mentioned backup, I get this:
root@newsauron:
Setting up slapd (2.4.7-6ubuntu3) ...
Backing up /etc/ldap/
Upgrading BDB 'checkpoint' options... .
Moving old database directories to /var/backups:
- directory dc=darkzone... done.
Loading from /var/backups/
- directory dc=darkzone... failed.
Loading the database from the LDIF dump failed with the following
error while running slapadd:
/var/
dpkg: error processing slapd (--configure):
subprocess post-installation script returned error exit status 1
Errors were encountered while processing:
slapd
When I remove both, I get this:
Setting up slapd (2.4.7-6ubuntu3) ...
Backing up /etc/ldap/
Upgrading BDB 'checkpoint' options... .
Moving old database directories to /var/backups:
Loading from /var/backups/
- directory dc=darkzone... failed.
Loading the database from the LDIF dump failed with the following
error while running slapadd:
/var/
dpkg: error processing slapd (--configure):
subprocess post-installation script returned error exit status 1
Errors were encountered while processing:
slapd
I have pulled an LDIF dump from the (still untouched) original server and rebuilt the database, so my LDAP is up and running again. How can I tell the postinst script that these actions are not necessary?
I have made a snapshot of the VM before starting the upgrade, and I'll investigate what went wrong after the first upgrade dry-run went through.
Ciao
Martin
I analyzed the problem(s), and I found invalid entries in my ldap database which seem to have been added during the installation a long time ago (See also Bug #112631).