LDAP troubleshooting

存储没打开IO读写cache所致。（此功能会差十几倍）

重新启动 ldap sever时, 出现下列错误信息 :

Checking configuration files for slapd: bdb_db_open: unclean shutdown detected; attempting recovery.
bdb_db_open: Recovery skipped in read-only mode. Run manual recovery if errors are encountered. config file testing succeeded

解决方法一 :

# service ldap stop
# /usr/sbin/slapd_db_recover -v -h /var/lib/ldap
# service ldap start 

解决方法二:

Description

The /etc/init.d/ldap try to stop the slapd process with the TERM signal first. If the process is active after three seconds, then the script stops slapd with the KILL signal. The next ldap start requires recovery:

# service ldap start Checking configuration files for slapd: bdb_db_open: unclean shutdown detected; attempting recovery. bdb_db_open: Recovery skipped in read-only mode. Run manual recovery if errors are encountered. --> hit Cntrl-C config file testing succeeded [OK ] Starting slapd: [OK ]

Solution

Using option -d with the killproc function. The delay time ca be configured with /etc/sysconfig/ldap. This solution allows a flexible configuration for the administrator.

Patch

This  and create the configuration file /etc/sysconfig/ldap with:

STOP_DELAY=60

Logfile

This example show a shutdown duration of 12 seconds:

Mar 11 19:23:50 slave slapd[9315]: daemon: shutdown requested and initiated. Mar 11 19:23:50 slave slapd[9315]: daemon: closing 7 Mar 11 19:23:50 slave slapd[9315]: daemon: closing 8 Mar 11 19:23:50 slave slapd[9315]: slapd shutdown: waiting for 1 threads to terminate Mar 11 19:23:50 slave slapd[9315]: =>do_syncrepl Mar 11 19:23:50 slave slapd[9315]: connection_get(12) Mar 11 19:23:50 slave slapd[9315]: connection_get(12): got connid=0 Mar 11 19:23:50 slave slapd[9315]: daemon: removing 12r Mar 11 19:23:50 slave slapd[9315]: slapd shutdown: initiated Mar 11 19:24:01 slave slapd[9315]: ====> bdb_cache_release_all Mar 11 19:24:02 slave slapd[9315]: slapd destroy: freeing system resources. Mar 11 19:24:02 slave slapd[9315]: slapd stopped.