Slapd dying.

[Adam Tauno Williams <adam@morrison-ind.com>]

I've got openldap-1.2.9-5 on a RedHat 6.2 system.  Slapd has developed the habit
of dying, in what seems a spontanious fashion,  that it I can't relate it to any
single query of event.  This is the networks "master" server, to which changes
are made,  but changes are rather infrequent, say, a dozen a day.  It is not
heavilly loaded as most queries go to various replicas,  all of which perform
fine.

I ran "/usr/sbin/slapd -d999" and sat back and waited for it to die.  Sure
enought, after about and hour and a half it died with:

====> cache_return_entry_r
entry_rdwr_runlock: ID: 871
User defined signal 1

What would cause this?  I'm certain nobody "kill"ed it as only three people can
logon to this box.  Openldap has been running for months without incident on
this box, and nothing significant has changed.

Systems and Network Administrator
Morrison Industries
1825 Monroe Ave NW.
Grand Rapids, MI. 49505