ldap startup failure with FC3

Craig White craigwhite at azapple.com
Tue Jan 3 03:27:29 UTC 2006


On Mon, 2006-01-02 at 21:42 -0500, Phil Savoie wrote:
> On January 2, 2006 20:24, Craig White wrote:
> > try chown -R ldap:ldap /var/lib/ldap
> >
> > add "loglevel 256" to slapd.conf
> >
> > echo "local4.*                  /var/log/slapd.log" > /etc/syslog.conf
> >
> > service syslog restart
> >
> > This will direct all ldap logging to /var/log/slapd.log so you can get a
> > better idea
> >
> > My guess is the first line will solve it...most ldap newbies run slapadd
> > as root and thus the data files are owned root:root and ldap will not
> > run.
> >
> > Craig
> 
> Hi Again,
> 
> Did what you asked and this is the resulted entries in the /var/log/slapd:
> 
> [root at server1 openldap]# tail /var/log/slapd.log
> Jan  2 21:30:48 server1 slapd[4404]: bdb(dc=example,dc=com): Ignoring log 
> file: /var/lib/ldap/log.0000000001: magic number 0, not 40988
> Jan  2 21:30:48 server1 slapd[4404]: bdb(dc=example,dc=com): Invalid log file: 
> log.0000000001: Invalid argument
> Jan  2 21:30:48 server1 slapd[4404]: bdb(dc=example,dc=com): PANIC: Invalid 
> argument
> Jan  2 21:30:48 server1 slapd[4404]: bdb(dc=example,dc=com): PANIC: 
> DB_RUNRECOVERY: Fatal error, run database recovery
> Jan  2 21:30:48 server1 slapd[4404]: bdb_db_open: dbenv_open failed: 
> DB_RUNRECOVERY: Fatal error, run database recovery (-30978)
> Jan  2 21:30:48 server1 slapd[4404]: backend_startup: bi_db_open(0) failed! 
> (-30978)
> Jan  2 21:30:48 server1 slapd[4404]: bdb(dc=example,dc=com): txn_checkpoint 
> interface requires an environment configured for the transaction subsystem
> Jan  2 21:30:48 server1 slapd[4404]: bdb_db_destroy: txn_checkpoint failed: 
> Invalid argument (22)
> Jan  2 21:30:48 server1 slapd[4404]: slapd stopped.
> Jan  2 21:30:48 server1 slapd[4404]: connections_destroy: nothing to destroy.
> [root at server1 openldap]#
> 
> Don't understand this magic number business...  Should I rm the log file and  
> touch another?  What db file is it referencing?  This is the listing 
> of /var/lib/ldap:
> 
> [root at server1 ldap]# ls -lR /var/lib/ldap
> /var/lib/ldap:
> total 20
> -rw-------  1 ldap ldap 460613 Jan  2 16:13 log.0000000001
> drwxr-xr-x  2 ldap ldap   4096 Jan  2 16:41 replica
> 
> /var/lib/ldap/replica:
> total 8
> -rw-r--r--  1 ldap ldap 0 Jan  2 16:41 slurpd.status
> -rw-r--r--  1 ldap ldap 0 Jan  2 16:43 slurpd.status.lock
> [root at server1 ldap]#
> 
> I don't see a db here but I guess there should be one?  As you can tell, I am 
> quit new to this.
----
the db will be there as soon as you slapadd/ldapadd some data

I would say that the best thing to do would be to stop ldap and remove
the log file and minimally populate it.

Are you using something as a reference?

if not, I would suggest using...

http://www.openldap.org/doc/admin22/

As you would need to populate it to have any db files

and I'm sorry about the recommendation to echo
echo "local4.*             /var/log/slapd.log" > /etc/syslog.conf # NO
which Alexander caught and should have been...
echo "local4.*             /var/log/slapd.log" >> /etc/syslog.conf #YES

I didn't however see any reference to checkpoints or logs in the
slapd.conf that you posted nor any references to DB_CONFIG files

Craig




More information about the fedora-list mailing list