RE: Slapadd with bdb backend segfaults

["Howard Chu" <hyc@highlandsun.com>]

> -----Original Message-----
> From: owner-openldap-bugs@OpenLDAP.org
> [mailto:owner-openldap-bugs@OpenLDAP.org]On Behalf Of Armbrust, Daniel C.

> We have been using openldap 2.1.22 with the ldbm backend on
> some fairly large databases.  Some of our databases seem to
> completely ignore the indexes when doing certain types of
> searches, causing very slow performance, and excessive memory
> usage - I may detail this issue in another e-mail.

back-ldbm's indexer has well-known limits; beyond those limits it is useless.
There is no need to report this in a separate email.

> Now, I am trying to load a ldif file with slapadd (that loads
> fine into ldbm with the same server) and it segfaults, after
> about 100 mb appear in the database directory.  I should also
> mention that with ldbm, this entire database loads in about
> 10 seconds... With bdb it takes about 3 minutes to get to the
> point where it fails.  If bdb databases are roughly the same
> size as ldbm databases, this means that it fails when it has
> it about 1/3 loaded.

Read the FAQ-o-Matic for information on properly configuring back-bdb. It
requires a bit more care than back-ldbm.
http://www.openldap.org/faq/index.cgi?file=893

> Questions:
>
> Is this a known issue?  Is there a fix?

See the OpenLDAP ITS for known issues. http://www.openldap.org/its/
If you don't find a similar report here, then your issue is not a known
issue.

> If not, when can I send to the list that would help debug the
> segfault?

A stack trace and the backend configuration from slapd.conf are a good start.

  -- Howard Chu
  Chief Architect, Symas Corp.       Director, Highland Sun
  http://www.symas.com               http://highlandsun.com/hyc
  Symas: Premier OpenSource Development and Support