Slapd pegs CPU every ten minutes

Hi folks -

I have an issue with oldap 2.36 in which every ten minutes, slapd starts consuming all available CPU resources briefly. When this happens ldap searches timeout.

I have bumped up the cache size to 1.5 GB dn2id.bdb and id2entry.bdb are approx. 60 mb combined.

I'm reading through this:

https://www.openldap.org/devel/admin/tuning.html

I get errors (mismatched version) when I run DB_STAT (database environment version mismatch) So, I don't have much more to report there.

I set logging back to the default to minimize logging.

There are three nodes in our cluster - one master.

Is there anything else I can look at to help troubleshoot? I am wondering what would cause slapd to peg the cpu at exactly 10 minute intervals.

Thanks for any help.

Norman Singley Directory Services 406 243 6799 Norman.singley@mso.umt.edu