Also update chart_version.sh script which
counts the number of commits since the BASE_VERSION
if BASE_VERSION tag is found. If the tag
doesn't exist it counts the number of commits since the
beginning. So when we use for example 2025.1.0 as
the BASE_VERSION but the tag 2025.1.0 is not yet created
the version of the e.g. nova chart will be calculated as
like 2025.1.563+<sha> and then when we create tag the nova version
will be 2025.1.0+<sha> which is undesired.
Lets use BASE_VERSION-<sha> if the tag is not found.
Change-Id: I032e8269ab17b490898d190adaec5c282e96fa88
Ceph 19.2.1 contains a critical RGW bug that can lose data. 19.2.2
is a hotfix release to address that issue.
This change updates the Ceph charts to new Ceph 19.2.2 images that
also contain a patch for another RGW bug that was previously
patched onto 19.2.1.
The Rook deployment is also updated to Rook 1.16.6 and is
configured to deploy Ceph 19.2.2.
Change-Id: I05ec9ed447e43a5b36f3004a84eb53cc846bc537
This commit adds the same tolerations that is in mariadb-server,
to the mariadb-controller pod definition.
Change-Id: Ida700bbb90bcbb06ac8de8a56899102494aa4c4d
Signed-off-by: Daniel Caires <DanielMarques.Caires@windriver.com>
This set of labels is not mandatory and introduces additional
complexity in upgrading existing Helm charts as labels field
is immutable and corresponding resources has to be deleted
before upgrade. The default behavior won't be changed and labels
won't be added only if explicitly disabled via values.
Change-Id: I2bc1cdefdc2819100195502003d69fd8576554ab
Signed-off-by: Ruslan Aliev <raliev@mirantis.com>
This is to align test scripts and playbooks with the recent
merger with the osh-infra repo.
* Get rid of OSH_INFRA_* env variables and
use OSH_* variables wherever necessary
* Update linter job so that playbook runs
the linting commands directly not involving tox
and additional shell script
* Move tools/gate/playbooks/* to playbooks/*
* Cleanup unused playbooks
* Cleanup unused shell scripts
Change-Id: I1913ed613025457d97cf1a4eeca6c2fbda0b72f6
Barbican moved to support multiple keks instead of just one.
OpenStack-helm is failing if run againist Barbican after commit [1].
This patch allows specify multiple and fix compatible issue.
[1] cfba1c1ba8f3659e6de727c3f1c274052e9ccace
Closes-Bug: 2103757
Change-Id: I438ee144b2a0a089dfffedf59961c155d9815889
"/" path causes readiness probe warning.
Change to use "healthcheck" api provided by manila.
story: 2011373
task: 51769
Change-Id: I6a720dd2863b6da256bd6f0466c5a2409333f5e9
* Add OVN support
* Add Octavia Driver Agent service
With this, we're allow to using OVN with Octavia just by adding OVN to
following configs:
* [api_settings]/enabled_provider_drivers
* [driver_agent]/enabled_provider_agents
Change-Id: I735c572fff80793806a9232c86e69059d0eda838
This change makes it possible to use identical
chart_version.sh script in both git repos
openstack-helm and openstack-helm-infra.
This is to prepare for the upcoming merger with
openstack-helm-infra and to reduce the number
of merge conflicts.
Change-Id: I18414cb0e45c0b595ba332363ef089bfbf37c5c0
This is to prepare for the upcoming merger with
openstack-helm-infra repo and reduce the number
of merge conflicts.
Change-Id: I26f73fb5116b45f0382a36605bd573355aa4fd03
This is to prepare for the upcoming merger
to the openstack-helm repo and to reduce
the number of merge conflicts.
Depends-On: I6a4166f5d4d69279ebd56c66f74e2cbc8cbd17dd
Change-Id: I3cb3f2c44d8401e1d0de673bf83f8e294433b8df
This is to prepare for the upcoming merger
with the openstack-helm-infra repo.
The doc files are copied from the openstack-helm-infra
repo w/o the commit history. The list of authors
is attached to the commit.
Co-Authored-By: Al Lau <al070e@att.com>
Co-Authored-By: Andreas Jaeger <aj@suse.com>
Co-Authored-By: astebenkova <astebenkova@mirantis.com>
Co-Authored-By: Chinasubbareddy Mallavarapu <cr3938@att.com>
Co-Authored-By: Gage Hugo <gagehugo@gmail.com>
Co-Authored-By: jinyuanliu <liujinyuan@inspur.com>
Co-Authored-By: John Haan <sjin.han@sk.com>
Co-Authored-By: Leontii Istomin <listomin@mirantis.com>
Co-Authored-By: lijunjie <lijunjie@cloudin.cn>
Co-Authored-By: Matthew Heler <matthew.heler@hotmail.com>
Co-Authored-By: Parsons, Cliff (cp769u) <cp769u@att.com>
Co-Authored-By: pengyuesheng <pengyuesheng@gohighsec.com>
Co-Authored-By: Q.hongtao <qihongtao@inspur.com>
Co-Authored-By: Roman Gorshunov <roman.gorshunov@att.com>
Co-Authored-By: Stephen Taylor <stephen.taylor.1@att.com>
Co-Authored-By: Steven Fitzpatrick <steven.fitzpatrick@att.com>
Co-Authored-By: Steve Wilkerson <sw5822@att.com>
Co-Authored-By: Steve Wilkerson <wilkers.steve@gmail.com>
Co-Authored-By: sunxifa <sunxifa@inspur.com>
Co-Authored-By: Tin Lam <tin@irrational.io>
Co-Authored-By: Tin Lam <t@lam.wtf>
Co-Authored-By: wangjiaqi07 <wangjiaqi07@inspur.com>
Change-Id: I6a4166f5d4d69279ebd56c66f74e2cbc8cbd17dd
This PR updates the yaml linter configuration
to align it with what we have in the osh-infra
repo.
Change-Id: I3585fdc6663b2ae7bfb5c1d8a13672ac3055bf86
Currently, the health-probe.py file opens connection
to amqp for rpc test, however, does not cleanly close
the client transport connection. This causes ampq like
rabbitmq to log warning messages and makes troubleshooting
difficult. This patch adds a clean transport.clenup once
the rpc test is over.
Change-Id: I55824d550344771368f97462836d5386cbf50fcd
When we deploy Ceph cluster using Rook we
check its status in a loop using Rook tools pod.
If tools pod is not found, let's skip the iteration.
Change-Id: Ib6bc90034961f89b8bb53081db5bf03c4d736110
It does not work on Ubuntu Noble which requires using virtual
env when you try to install packages using pip. Also
for deployment tests reno is not needed because we build
charts with SKIP_CHANGELOG=1
Change-Id: I8f0578ed2e1d0e757add155c618eea2e8a2e30d2
It fails with the error: `No module named 'yaml'`.
So let's use native helm with the ansible shell instead.
Change-Id: If652d603cfcaeb0b70c9b566b90d98e627d3bada
metallb isn't really a feature gate because
it is not related to any override. Instead we
used it only during ingress deployment to decide
whether we need to deploy cluster ingress instance.
Previously we moved ingress deployment to the deploy-env
role and cluster ingress instance is deployed
when metallb_setup job variable is set.
Change-Id: I22f6864e40f86527826a40379e516228f43896da
While we are waiting for Ceph cluster to be ready
we check Ceph status in a loop using tools pod provided by Rook.
We have to get this tools pod name every iteration
within the loop because K8s can terminate the pod for
some reason and this is expected behavior.
Change-Id: Iabb98e94d7470fe996091bf77787637f3e8f4798
A while ago we changed the way how we define features
for the test jobs and this variable is not needed any more
Also change keystone jobs so they don't deploy
Rook but instead use local volumes.
Depends-On: Ib4afe58b27cd255ce844626b1eee5ecc82e3aeb3
Change-Id: Ia8971bd8c3723542a275c9658db7f9a5bb943f92
* Make TLS job non-voting because it is unstable and
we don't want it to block the gate.
* Ceph migration job is not very important now. Let's
run it in a periodic-weekly pipeline.
Change-Id: Iadb67e1c5218794d15e60538abb2e869ae7e67c0
* We recently removed resource limitations
for Rook deployment, so it can use recommended
defaults. However it requires more memory.
This PR updates jobs so all jobs that deploy Rook
use the 5 nodes nodeset.
* Also MetalLB deployments will use 1+3 nodes nodesets.
Change-Id: Ia5f5d436e1503e11833bbb9b014807a911541406
