ITS#1986

[stroh@scan-plus.de]

the error occurs when slapd has run for some time, I've never seen it
just after restart of the daemon:

# ldapadd -D "<amin dn>" -x -f <ldif file>
adding new entry "<dn of the new entry>"
ldap_add: Unknown error
        additional info: cache add failed
ldif_record() = 80

some facts:
slapd version: 2.0.23-6 (compiled with SSL support)
OS: debian-woody linux, kernel 2.4.18, filesystem is XFS
the number of indices is extremly high[1], maybe that's one of the main
problems. The database is holding about 25.000 records at this moment
(structured, max. 100 records per subtree-level).

[1]:
index default pres,eq
index objectClass eq
index uidNumber,gidNumber eq
index ServerName eq
index uid,contactListId eq
index mail,mailAlternateAddress,mailForwardingAddress eq
index cn,sn,firstName eq,sub
index corpName,corpUnit,corpTitle,corpStreet,corpLocation eq,sub
index privStreet,privLocation,privPostalCode,corpPostalCode eq,sub
index street,localityName,postalCode eq,sub
index telephoneNumber,facsimileTelephoneNumber eq,subinitial
index privTelephoneNumber,privFacsimileTelephoneNumber eq,subinitial
index corpTelephoneNumber,corpFacsimileTelephoneNumber eq,subinitial
index contactCategory pres,eq
index accountStatus eq

regards,
Igor Stroh