collectd is not reporting platform memory properly

[Frank Miller]

Brief Description
-----------------
The collectd memory collection is not properly splitting out the memory usage between the platform and applications.

Severity
--------
Minor

Steps to Reproduce
------------------

Expected Behavior
------------------
Would like to see collectd report memory broken down by platform and applications. For the platform memory would like to see this further broken down by kubernetes system processes vs base (eg: flock services) processes vs stx-monitor or stx-openstack processes.

Actual Behavior
----------------
collectd platform memory is not currently accurate. Also collectd platform memory does not separate out the key platform components (kubernetes, flock, stx-openstack, stx-monitor).

Reproducibility
---------------
Reproducible

System Configuration
--------------------
All configs

Branch/Pull Time/Commit
-----------------------
Any stx.3.0 or earlier load

Last Pass
---------
n/a

Timestamp/Logs
--------------
n/a

Test Activity
-------------
System testing

[Ghada Khalil (gkhalil)]

stx.3.0 / medium priority - tool limitation impacting debugging

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/689746

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/689749

[OpenStack Infra (hudson-openstack)]

Change abandoned by Jim Gauld (<email address hidden>) on branch: master
Review: https://review.opendev.org/689746
Reason: No longer need this change, will change monitoring memory plugin to parse existing worker_reserved.conf WORKER_BASE_MEMORY.

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/689749
Committed: https://git.openstack.org/cgit/starlingx/monitoring/commit/?id=6bb5be1a65ff7f9c65da66f28566c654f31cb291
Submitter: Zuul
Branch: master

commit 6bb5be1a65ff7f9c65da66f28566c654f31cb291
Author: Jim Gauld <email address hidden>
Date: Mon Oct 21 09:13:51 2019 -0400

    Update collectd breakdown of platform memory

    This updates collectd memory metrics. New metrics are dispatched and
    logged to give a better platform breakdown.
    - 4K memory usage is split between platform and applications.
    - Platform memory metric calculation is updated to be percent of
      platform memory usage out of platform reserved memory.
    - Platform memory is broken down into base, kube-system, and addon.
    - The 4K memory is broken down into anon, avail, total buckets.

    This also adds verbose logging of these metrics per memory sample
    collection audit.

    The following collectd samples are dispatched with these headings:
    type, type_instance, plugin, plugin_instance: description

    percent, used, memory, platform:
     platform memory usage out of platform reserved memory (%)

    absolute, used, memory, anon:
     anonymous 4K memory used based on meminfo (MiB)

    absolute, used, memory, avail:
     available 4K memory used based on meminfo (MiB)

    absolute, used, memory, total:
     total 4K memory pool based on meminfo (MiB)

    absolute, used, memory, cgroup-rss:
     overall 4K memory used based on cgroups (MiB)

    absolute, used, memory, platform:
     platform reserved memory (MiB)

    absolute, used, memory, platform:
     platform memory usage(MiB)

    absolute, used, memory, base:
     base memory usage (MiB)

    absolute, used, memory, kube-system:
     kubernetes kube-system namespace memory usage (MiB)

    absolute, used, memory, kube-addon:
     kubernetes monitor and openstack namespace memory usage (MiB)

    Verbose memory logging may be enabled manually by setting
    "verbose = true" in /etc/collectd.d/python_plugins.conf configuration
    file and restarting collectd service on the node.

    Closes-Bug: 1848751
    Change-Id: Ibacd24287d9f585b843adb7da7df6cc09961be42
    Signed-off-by: Jim Gauld <email address hidden>