Comment 13 for bug 1949360

Reviewed: https://review.opendev.org/c/starlingx/openstack-armada-app/+/817492
Committed: https://opendev.org/starlingx/openstack-armada-app/commit/c28605b14f38927dedcef0a543261afade1441a4
Submitter: "Zuul (22348)"
Branch: master

commit c28605b14f38927dedcef0a543261afade1441a4
Author: Delfino Curado <email address hidden>
Date: Wed Nov 10 17:51:27 2021 -0500

    Fix ceph pool creation during stx-openstack apply

    Previously the helm overrides for nova, cinder and glance respectively
    defined 512, 256 and 256 as the PG_NUM for their Ceph pools. Until
    Mimic, ceph would just issue a warning message if this number was
    bigger than the number OSD * 100, now Nautilus returns an error message
    and asks for expected_num_objects parameter or --yes-i-really-mean-it.
    Neither of these options is supported by openstack-helm.

    The pool creation is adjusted to take into account the number of OSDs
    available to choose the PG_NUM.
    The steps are:
    Number of OSD times 100 minus 1;
    Get the nearest power of two numbers below this result;
    Limit PG_NUM to the previous defaults, as they are already high
    numbers.
    This logic roughly results in the same values described here:
    https://docs.ceph.com/en/nautilus/rados/operations/placement-groups/#a-preselection-of-pg-num

    This was done to solve an error message which demands a
    expected_num_objects when PG_NUM is considered too high by ceph.

    Test plan:
    StarlingX builds successfully
    stx-openstack is built successfully
    stx-openstack is applied successfully and ceph pools are created
    accordingly
    It is possible to override PG_NUM on ceph pools by changing chunk_size
    value through helm overrides

    Closes-Bug: #1949360
    Depends-On: I222bee29bcaa09a95a3706c72dd21b8ed3efbe60
    Signed-off-by: Delfino Curado <email address hidden>
    Change-Id: Ia1416e64afcdf91b86afdf750bf5b3a1727db985