Re: (ITS#4242) slapindex fails to clean up DB after unclean shutdown

[hyc@symas.com]

This works as designed. Since transaction logging is disabled when using 
-q, it is not possible to recover the database to a consistent state 
when these tools get interrupted. I.e., you are supposed to (A) never 
interrupt a -q run, and (B) start over from scratch (rm *) when you do 
interrupt the run.

quanah@stanford.edu wrote:
> Full_Name: Quanah Gibson-Mount
> Version: 2.3.13
> OS: Solaris 8
> URL: ftp://ftp.openldap.org/incoming/
> Submission from: (NULL) (171.66.155.86)
>
>
> I cancelled a slapindex that was taking forever to make some configuration
> adjustments.  After doing this, I started the new one, which failed:
>
> ldap9:/root# time slapindex -q
> bdb_db_open: unclean shutdown detected; attempting recovery.
> bdb_db_open: Database cannot be recovered. Restore from backup!
> backend_startup_one: bi_db_open failed! (-1)
> slap_startup failed
> 0.21u 0.04s 0:00.21 119.0%
> ldap9:/root# time slapindex -q
> bdb_db_open: unclean shutdown detected; attempting recovery.
> bdb_db_open: Database cannot be recovered. Restore from backup!
> backend_startup_one: bi_db_open failed! (-1)
> slap_startup failed
> 0.20u 0.04s 0:00.21 114.2%
>
>
> I finally removed alock, and then it worked:
>
> ldap9:/db# \rm alock
> ldap9:/db# cd
> ldap9:/root# time slapindex -q
>
>
> --Quanah
>
>
>
>
>   


-- 
  -- Howard Chu
  Chief Architect, Symas Corp.  http://www.symas.com
  Director, Highland Sun        http://highlandsun.com/hyc
  OpenLDAP Core Team            http://www.openldap.org/project/