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

database corruption when batch processing (ITS#2498)



Full_Name: Kevin Reichhart
Version: 2.1.4
OS: SuSE 8.0
URL: ftp://ftp.openldap.org/incoming/
Submission from: (NULL) (64.80.42.147)


When running a script to do batch ldapadd or ldapmodify, the database becomes
corrupt, the ldapadd/ldapmodify hangs indefinately and slapd hogs 99-100% of one
or both CPUs.  Stopping and starting ldap has no effect - as soon as ldap is
restarted, slapd immediately hogs the CPU(s).  I have also seen this on a RedHat
8.0 box.  I'm running sleepycat (bdb) in the back.  The only fix I've found is
to replace the database with a backup and start over, hoping for better results.
 I have tried putting a sleep() in my script (from 1 to 3 seconds) and no
difference.