Chart upgrading was failing due to some immutable fields in job are needed to upgrade. So, we thought using the post-install and post-upgrade
helm hook for job to force the job resource to execute after all resources are created. And as some jobs are dependent on each other i.e.
some jobs needs to run in order for helm hook to be successful. For that we used hook-weight to control resource creation order.
Change-Id: I26881324d101a986b7367d4682e9adcd07a24b13
It is possible than an interface has multiple IP addresses, for
simplicity of this change, use the first one so that the service
can start.
We can look later into improving it to accept some sort of index
for the IP address.
Change-Id: Ie856f54331d689a51bfd6de45db5820b765797ef
There was an issue with the metadef APIs in glance, detailed in
the latest OSSN[0] that they have the potential to leak resources.
This change updates the default policy for the metadef APIs to
be disabled by default.
[0] https://wiki.openstack.org/wiki/OSSN/OSSN-0088
Change-Id: I7377b3a2f3784fe7da78bdd7aba146328cc0f406
Verification is added to Cinder volume type creation logic
under Values.bootstrap to make sure the volume_backend_name is
configured in Values.conf.backends.
Change-Id: I1b9b1eaac8df861d28d4121477de148dba6a2dbf
When using a helm3 to deploy , it fails. Helm3 no more support rbac.authorization.k8s.io/v1beta1 , but v1 can support helm2 and helm3.
Change-Id: If37ec26443feb5328d49e6b3c419305832bdae9e
Nova will check if pty device exist or not under /dev/pts
when get console log.
If it does not exist, cannot get console log.
ref: https://review.opendev.org/c/starlingx/config/+/660268
Change-Id: I2793d1f51c18e81a4271b8b0c50bfe1a2dab8a09
When using a helm3 to deploy , it fails. Helm3 no more support rbac.authorization.k8s.io/v1beta1 , but v1 can support helm2 and helm3.
Change-Id: Ie4e1c79c2b2513318d1b7d6a13712921a6c2a6cb
This change updates the releasenotes for the horizon chart to the latest version of the chart as of this patchset.
Change-Id: I7b2df90340f33a5b3edac5a53d8031325f96244d
When Deployment of compute nodes is not on all nodes (e.g.Total 5 nodes,but 3 compute nodes),The original method counts all nodes instead of compute nodes,This can result in less than 100% and the process will get stuck,this is a bug!
Change-Id: I39c5d2014146925afe7fd896123a705c19005ff9
With a previous change[0] that moved rabbit-init jobs to dynamic
in helm-toolkit, this change continues that work by moving the
keystone rabbit-init job to dynamic as well.
[0] https://review.opendev.org/c/openstack/openstack-helm-infra/+/671727
Change-Id: Iec2ea3fdf36e19ac4f2e203389dbe19737d14c3a
