Re: hot backups for openldap bdb backend.

[Jehan PROCACCIA <Jehan.Procaccia@int-evry.fr>]

what do you mean by hot ? while slapd is running ? i had problems doing 
that !

in terms of backup, I just know that the following did worked (with 
slapd down !) for me once:

[root@corbeau /var/lib/ldap/int]
$ /usr/sbin/slapd_db_archive -s -v
IntEPersInetServ.bdb
cn.bdb
dn2id.bdb
gidNumber.bdb
givenName.bdb
id2entry.bdb
mail.bdb
objectClass.bdb
sn.bdb
uid.bdb
uidNumber.bdb
[root@corbeau /var/lib/ldap/int]
$ /usr/sbin/slapd_db_archive -v
log.0000000001
log.0000000002
log.0000000003
log.0000000004
log.0000000005
log.0000000006
log.0000000007

[root@corbeau /var/lib/ldap/int]
$ cp log.0000000008 ../int.sav/
[root@corbeau /var/lib/ldap/int]
$ cp IntEPersInetServ.bdb cn.bdb dn2id.bdb gidNumber.bdb givenName.bdb
id2entry.bdb mail.bdb objectClass.bdb sn.bdb uid.bdb uidNumber.bdb 
../int.sav/
[root@corbeau /var/lib/ldap/int]
$ rm *

$ cp ../int.sav/* .
[root@corbeau /var/lib/ldap/int]
$ ls
cn.bdb     gidNumber.bdb  IntEPersInetServ.bdb  objectClass.bdb 
uidNumber.bdb
DB_CONFIG  givenName.bdb  log.0000000008        sn.bdb
dn2id.bdb  id2entry.bdb   mail.bdb              uid.bdb
You have new mail in /var/spool/mail/root
$ chown ldap:ldap *
$ /etc/init.d/ldap start

Actually I'am more familiar with a trditional ldif (slapcat) of the 
database ... but if there's a better way to do it, please let me know .

Peter Johnson wrote:
> Hi,
>
> 	I'm tring to do hot backups of my openldap server using db_archive and 
> db_recover.   Whenever I try using either of these commands I get a 'log 
> record size mismatch' error.    What am I doing wrong?  Is there a mailing 
> list for bdb?  I didn't see on on the sleepycat.com site.
>
>
> -- [oracle@server3837 openldap-reload]# 
> /usr/local/BerkeleyDB.4.1/bin/db_archive -l -P xxxxxxx -v
> db_archive: log record size mismatch
> db_archive: DB_LOGC->get: invalid log record header
> db_archive: PANIC: Input/output error
> db_archive: PANIC: DB_RUNRECOVERY: Fatal error, run database recovery
> db_archive: open: DB_RUNRECOVERY: Fatal error, run database recovery
>
>
> Thanks
>
> Peter Johnson	
> Humboldt State University
> 707-826-6122



--
Jehan Procaccia 			| Ingenieur Systemes & Reseaux
Institut National des Telecommunications| Tel : +33 (0) 160764436
MCI, Moyens Communs Informatiques  | Mail: Jehan.Procaccia@int-evry.fr
9 rue Charles Fourier 91011 Evry France | Fax : +33 (0) 160764321