This content has been marked as final. Show 6 replies
Are you running checkpoints, because logs cannot be removed until a checkpoint is executed? For more information read http://docs.oracle.com/cd/E17076_02/html/gsg_txn/C/logfileremoval.html
Yes we are running checkpoints
Our current DB_CONFIG are:
set_cachesize 0 268435456 1
we use slapd which runs checkpoint after every 0 bytes with 2 minutes
I see your problem, you are using the wrong command in DB_CONFIG to set DB_LOG_AUTO_REMOVE. To set it use:
log_set_config DB_LOG_AUTO_REMOVE 1
I guess log_set_config is the new syntax for berkely db . We are using Sleepycat Software: Berkeley DB 4.4.20: which is an embeded db for OpenLDAP
Checking configuration files for slapd: bdb(dc=example,dc=com): unrecognized name-value pair: log_set_config
bdb_db_open: Database cannot be opened, err 22. Restore from backup!
bdb(dc=example,dc=com): DB_ENV->lock_id_free interface requires an environment configured for the locking subsystem
backend_startup_one: bi_db_open failed! (22)
slap_startup failed (test would succeed using the -u switch)
Can you let us know what are chances of a log file not getting removed even though db_archive -l shows that log file is removable and not in use when .
db_archive -l prints out the names of all the log files, not just those that are ready to be removed. If you want just a list of removeable logs use "db_archive", without the -l argument.
Are the logs printed out by "db_archive" eventually getting removed? If not, does using "db_archive -d" remove those log files?