[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: db corruption?

I got no response to my previous question regarding deleting replogs, rej
logs and status files, but I believe I've figured it out through trial
and error.  An affirmation of those beliefs would be nice though.

I took down the master slapd and slurpd, then took down the slave slapd.
I copied the .dbb files to the slave slapd, removed the .rej logs, the
.status log and the .replog log.  I then brought services back up. Things
seemed to be ok for a while, but my DBs have drifted out of synch again.

If I run an ldapsearch on the replica for one of the affected accounts, I
get no results.  I try to delete the account, and I get:

Sounds like database corruption to me. Did you have an unclean shutdown? Have you tried running db_recover? Have you thought of using slapcat on the master and then slapadd on the replica?


I don't have db_recover. Is that a script or an ldap-util? My plan for fixing it tonight is to do the slapcat and then slapadd on the replicas, followed by slapindex on the replicas.

On the master server, all queries work ok, and the accounts in question will successfully add or delete depending on what I'm trying to do. I am worried though, that after two attempts at simple file copy, the master DB might be corrupted? If that's the case, would I slapcat out the contents, back up the db, and then try to rebuild it with slapadd and slapindex?

Thanks a lot Quanah

Aaron Thoreson
Network Group
Midcontinent Communications