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

Re: keep getting errors on this DSA



On Tue, 2005-02-15 at 03:06 -0800, Howard Chu wrote:
> Craig White wrote:

> >
> You moved your DB_CONFIG file into place after the BDB environment was 
> already created by slapadd, and then you started slapd. Apparently your 
> non-default settings include putting the logs in a non-default location. 
> Naturally the log settings that slapd sees disagree with the log 
> settings in effect during the slapadd, so slapd is unable to find any 
> valid logs.
> 
> You're supposed to set the DB_CONFIG first so that slapadd and slapd 
> both operate with identical settings.
-----
This makes some sense to me - it's funny how I have the same setup at my
house and have not experienced any problem whatsoever. I moved DB_CONFIG
file deleted all the data files and the log files, and moved DB_CONFIG
back.

I then did the slapadd to reload the db, changed ownership to ldap:ldap
*bdb and __db* and then started slapd again. Stopped slapd, checked with
db_recover and no problems (progress), went to add a single entry and
failed with the following error...

adding new entry "ou=AddressBook,dc=tobyhouse,dc=com"
ldap_add: Internal (implementation specific) error (80)
        additional info: txn_commit failed

This is the problem I've been having all along and I obviously have to
get to the bottom of it.

# cat DB_CONFIG
#
# Set the database in memory cache size.
#
set_cachesize   0       52428800        0

# Automatically remove log files that are no longer needed.
set_flags DB_LOG_AUTOREMOVE

#
# Set database flags.
# (for database loading/reindexing)
#set_flags       DB_TXN_NOSYNC
#set_flags DB_TXN_NOT_DURABLE

# Set log values.
#
set_lg_regionmax        1048576
set_lg_max              10485760
set_lg_bsize            2097152
set_lg_dir              /usr/local/var/openldap-logs

and perhaps the relevant section from slapd.conf

# tail /usr/local/etc/openldap/slapd.conf
# Set the entry cache size to 5000.
#
cachesize       5000

# Set transactional checkpoint.
#
checkpoint      512     720

Why am I having so much trouble on this system but not at home with the
same settings? (Note: I have left the 'set_flags' in DB_CONFIG commented
out throughout my efforts)

Not wishing to have 'non-settings' putting log files in 'non-default'
locations, I am amenable to any suggestions or further clarity anyone
may provide to this problem. 

Craig