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

Re: slapd crashes (ldap 2.1.30, db 4.1.25)

Hash: SHA1

John Borwick wrote:
| We added "--enable-rewrite" to the spec file and rebuilt the RPM.  We
| also created a DB_CONFIG file:
| set_cachesize 0 536870912 1
| set_lg_regionmax 262144
| set_lg_bsize 2097152
| Our server's been running for 18 hours now, without a crash.  I'll
| follow up if we start getting more crashes.

OK.  OpenLDAP crashed on us three more times yesterday, starting a few
hours after I sent this message.

Our priority is to get LDAP running in a production capacity _right
now_.  I just can't figure out what's making LDAP crash.  Sometimes it
crashes in an hour when a few people are logged in, sometimes it crashes
when there are 500 people querying, sometimes it crashes after 8 hours
when it's been hit hard and then sitting idle.  The queries look normal

What category of thing could be causing the crash?  We've brainstormed
~   * backend rewrite of o=WFU,c=US to ou=Users,dc=wfu,dc=edu
~        - already recompiled with "--enable-rewrite"
~   * number of total connections
~        - unlikely, seems to vary widely
~   * number of open connections
~        - unlikely but possible
~   * database corruption
~        - no complaints in the error log
~   * database version
~        - Using BerkeleyDB 4.1.25 as Red Hat ships
~   * OS
~        - Red Hat Workstation 3
~   * replication
~        - no errors in the master's LDAP log
~   * network problems

Thanks for all your help!
- --
~            John Borwick
~        Systems Administrator
~       Wake Forest University | web  http://www.wfu.edu/~borwicjh
~       Winston-Salem, NC, USA | GPG key ID               56D60872

Version: GnuPG v1.2.4 (FreeBSD)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org