Description of problem: Pre-flight storage checks do not go into enough detail with respect to custom partitioning. https://github.com/openshift/openshift-ansible/blob/release-3.6/roles/openshift_health_checker/openshift_checks/disk_availability.py The "etcd" node precheck looks for 20Gb of free space on /var. In the case where /var represents multiple partitions, the check fails. For example: /var is 10Gb in size but as a separate partition, /var/lib/etcd is the required 20Gb in size. It should be possible to specify or check for a more specific directory/filesystem layout for required space. Version-Release number of the following components: [root@osemaster1 ~]$ rpm -q openshift-ansible openshift-ansible-3.6.173.0.21-2.git.0.44a4038.el7.noarch [root@osemaster1 ~]$ rpm -q ansible ansible-2.2.3.0-1.el7.noarch [root@osemaster1 ~]$ ansible --version ansible 2.2.3.0 config file = /etc/ansible/ansible.cfg configured module search path = Default w/o overrides How reproducible: Steps to Reproduce: 1. Have a mount point for a required directory, like /var/lib/etcd, with the correct recommended size 2. Have the top level that is checked, like /var, less than the recommended size 3. Run precheck Actual results: Precheck will fail on storage sizing Expected results: Should be able to account for or tolerate different layouts. Additional info:
This check has been fixed in 3.6 to work as you say. https://access.redhat.com/errata/RHBA-2017:3438 contained the fix for https://bugzilla.redhat.com/show_bug.cgi?id=1491566 -- all sub-mounts of /var are summed to meet the threshold. *** This bug has been marked as a duplicate of bug 1491566 ***