Just want to share that is known issue to our cluster:

1 - install new replica

2 - install of the replica fails for any reason (in my case it was due to I am unable to set the server which custodia uses in the ipa-server-istall command line)

3 - ipa-server-install --uninstall

4 - RUVs from 1970--00-00 and slapd's eating all the CPU they have.

So correct way is always to clean up the failed replica from the cluster first, not to use --uninstall. It's centos 7.5.

--