公司的一台 LDAP server (FreeBSD)傍晚時因為某些原因所以緊急重開，重開後發現 slapd 跑不起來。
# /usr/local/etc/rc.d/slapd start /usr/local/etc/rc.d/slapd: WARNING: failed to start slapd
原本以為可能是 ram 出問題導致靈異，仔細一看又不太像。
用 -d 下去跑發現…
# /usr/local/libexec/slapd -u ldap -g ldap -d -1 ... ... (中間省略) ... ... 5228a6a1 bdb_db_open: database "dc=XXX,dc=com": dbenv_open(/var/db/openldap-data). 5228a6a1 bdb_db_open: database "dc=XXX,dc=com" cannot be opened, err 22. Restore from backup! 5228a6a1 ====> bdb_cache_release_all 5228a6a1 bdb(dc=XXX,dc=com): txn_checkpoint interface requires an environment configured for the transaction subsystem 5228a6a1 bdb_db_close: database "dc=XXX,dc=com": txn_checkpoint failed: Invalid argument (22). 5228a6a1 backend_startup_one (type=bdb, suffix="dc=XXX,dc=com"): bi_db_open failed! (22) 5228a6a1 slapd shutdown: initiated 5228a6a1 ====> bdb_cache_release_all 5228a6a1 bdb_db_close: database "dc=XXX,dc=com": alock_close failed 5228a6a1 slapd destroy: freeing system resources. 5228a6a1 slapd stopped.
看起來就是 ldap db 爛了…
用 db_recover 來試著修修看（不同版本 bdb 的 db_recover 代不同版號），修的好就沒問題，修不好就得從備份撈
# /usr/local/bin/db_recover-4.6 -v -h /var/db/openldap-data/ Finding last valid log LSN: file: 1 offset 4541603 Recovery starting from [1][4541458] Recovery complete at Thu Sep 5 23:48:00 2013 Maximum transaction ID 800000ec Recovery checkpoint [1][4541603]
然後再跑一次 slapd
# /usr/local/etc/rc.d/slapd start Starting slapd.
就跑起來啦~~