This patch set adds efficacy indicators for workload_balancing
goal (that includes workload_stabilization and workload_balance
strategies so far).
Change-Id: I5b04d084ace7c661001c62f07b8308e5763e144d
This patch set refactors logs of workload stabilization
strategy to make them more readable and sensible.
Change-Id: I408988712bb7560728157f3b4e4f2b37572128c4
This patch set adds background job that reschedules CONTINUOUS
audits from FAILED to ACTIVE decision engine nodes using round-robin
algorithm. It also contains fix for main[1] HA PS about filtering audits.
[1]: https://review.openstack.org/#/c/578102/
Partially-Implements: blueprint support-watcher-ha-active-active-mode
Change-Id: Ib248a6cd3adbd3927c47db6bb819300361492411
This patch set adds hostname field to Audit and Action Plan
objects to track services which execute these objects.
Change-Id: I786e419952925c380c969b12cc60f9a1004af96b
Partially-Implements: blueprint support-watcher-ha-active-active-mode
This patch replaces voluptuous with JSON-schema to validate
efficacy indicator since in watcher we want to remove voluptuous
and use JSONSchema as our only JSON validation tool to keep consistency.
Change-Id: Iaa77566f1cdfdac03ce8e7d5a75406274c7d5298
Implements: blueprint replace-voplutuous-with-jsonschema
The python 3 version of the linter does not allow Windows-style line
endings (\r\n) so replace them with UNIX-style endings (\n).
Change-Id: Ifb97491323d3df92bb1520e373552aeb5e1919a4
Signed-off-by: Doug Hellmann <doug@doughellmann.com>
Storage CDM can not be build for some environment such as
the one using VMwareVcVmdkDriver, since some attributes of
Storage CDM'S pool element can be 'unknown'.
This patch updates storage CDM collector to raise watcher
specific exception if some attributes of storage CDM'S pool
element is 'unknown'
Change-Id: If75a909025c8d764e4de6e20f058b84e23123c1a
Closes-Bug: #1751206
maintain one compute node without having the user's application
been interruptted.
It will firstly migrate all instances from the maintenance node
to one backup node. If not, it will migrate all instances,
relying on nova-schduler.
Change-Id: I29ecb65745d5e6ecab41508e9a91b29b39a3f0a8
Implements:blueprint cluster-maintaining
Currently RabbitMQ connection gradually increases by CONTINUOUS audit
with auto-trigger option.
This patch fixes watcher to reuse RabbitMQ connection.
Change-Id: I818fc1ce982f67bac08c815821f1ad67f8f3c893
Both Continuous and Oneshot audits made same action in
do_execute, so it's a good idea to move it to the base
class
TrivialFix
Change-Id: Ic0353f010509ce45f94126e4db0e629417128ded
This patch adds project_id in compute CDM, It also adds logic for
excluding project_id in audit scope.
Change-Id: Ife228e3d1855b65abee637516470e463ba8a2815
Implements: blueprint audit-scope-exclude-project
Removed duplicates of strategies descriptions, added references to
that descriptions instead of module descriptions.
Change-Id: Ife396ddce5c3cc926cc111f1ff1abd3a42c22561
NotImplementedError are reported in decision-engine log file
when we activate storage data model and see a Guru Meditation Report.
This patch fixes by adding default values.
Change-Id: I06386f8295f7758cbb633612eee8b19225905c92
Closes-Bug: #1750300
This patch set completes schema by adding restrictions
to different types of schema properties.
It also makes workload_stabilization strategy more
user friendly by setting cpu_util as default metric.
Change-Id: If34cf4b7ee2f70dc9a86309cb94a90b19e3d9bec
