(ITS#3970) slapindex fails after clean database

[quanah@stanford.edu]

Full_Name: Quanah Gibson-Mount
Version: 2.3.6
OS: Solaris 8
URL: ftp://ftp.openldap.org/incoming/
Submission from: (NULL) (171.64.19.82)


I found that in the following scenario, slapindex will complain about an unclean
database shutdown:

Stop slapd
Update an index
run slapindex -q
(fails)
run slapindex -q
(fails)
run slapindex -q
succeeds

without debugging, it shows:

ldap-dev2:/root# time slapindex -q
slap_startup failed
0.28u 0.05s 0:00.34 97.0%
ldap-dev2:/root# time slapindex -q
slap_startup failed
0.31u 0.07s 0:00.38 100.0%
ldap-dev2:/root# time slapindex -q

With -d -1 the error is:

backend_startup_one: starting "dc=stanford,dc=edu" 
  bdb_db_open: dc=stanford,dc=edu 
  bdb_db_open: unclean shutdown detected; attempting recovery. 
  bdb_db_open: dbenv_open(/db) 
  bdb_db_open: recovery failed: No such file or directory (2) 
  backend_startup_one: bi_db_open failed! (2) 

I've been able to reproduce this multiple times.

--Quanah