magnum/magnum
Bharat Kunwar 55fd12a47c [fix] Use default_ng_worker.node_count for patches
At present, when a cluster is patched without node_count (e.g. when
health_status and health_status_reason are patched), the node_count in
the cluster_update invocation contains cluster.node_count instead of
default_ng_worker.node_count which causes unexpected behaviour when a
cluster has additional nodegroups (e.g. the default-worker size gets
incremented by 1 if the additional nodegroup has 1 worker each time the
cluster is patched).

This PS fixes this issue by using default_ng_worker.node_count if path
"/node_count" is not present in the patch object.

Story: 2005266
Task: 39962

Change-Id: I0df5149860604f0adf52701df371e34a6e667b9d
2020-06-11 10:30:21 +00:00
..
api [fix] Use default_ng_worker.node_count for patches 2020-06-11 10:30:21 +00:00
cmd Monkey patch original current_thread _active 2020-05-05 09:39:35 -04:00
common Merge "Support upgrade on behalf of user by admin" 2020-05-26 11:19:19 +00:00
conductor [k8s] Support updating k8s cluster health status 2020-04-09 16:59:26 +12:00
conf [k8s] Support configurable health polling interval 2020-05-16 12:15:50 +12:00
db Update hacking for Python3 2020-03-31 20:09:46 +02:00
drivers Merge "[hca] Pin fedora to 32 until new greenlet release" 2020-06-03 10:28:14 +00:00
hacking Update hacking for Python3 2020-03-31 20:09:46 +02:00
objects Improve log of k8s health status check 2019-10-01 18:29:16 +00:00
service [k8s] Support configurable health polling interval 2020-05-16 12:15:50 +12:00
servicegroup Trivial code cleanups 2018-10-02 19:41:34 +00:00
tests [fix] Use default_ng_worker.node_count for patches 2020-06-11 10:30:21 +00:00
__init__.py Initial commit from github (squashed) 2014-11-18 09:23:37 -05:00
i18n.py Remove log translations 2017-03-30 17:05:10 +08:00
version.py Remove coding:utf-8 2015-08-03 23:28:22 +08:00