v0.20.0

v0.20.0

Changes

Enhancement

• Add OpenShift-compatible Dockerfiles. (#509, @bcrochet)
• Allow make cluster-sync on external systems. (#500, @qinqon)
• Add NodeSelector optional field to operator spec, to control on which nodes the handler should run on. (#514, @yboaron)

Other (Bug, Cleanup or Flake)

• Fix the name from NMstate to NMState. (#521, @phoracek)
• Metrics were disabled to avoid conflicts of TCP ports and reduce resource requirements. (#518, @phoracek)
• Only one NMstate.nmstate.io CR should be allowed. This change will ignore any CR that is not named nmstate. (#512, @bcrochet)
• Webhook runs now as unprivileged Deployment
  action requiered: Wait for webhook Deployment Ready status too after deploy/upgrade. (#497, @qinqon)

Uncategorized

• Make the operator's API beta. (#523, @bcrochet)

Installation

First, install kubernetes-nmstate operator:

kubectl apply -f https://github.com/nmstate/kubernetes-nmstate/releases/download/v0.20.0/nmstate.io_nmstates_crd.yaml
kubectl apply -f https://github.com/nmstate/kubernetes-nmstate/releases/download/v0.20.0/namespace.yaml
kubectl apply -f https://github.com/nmstate/kubernetes-nmstate/releases/download/v0.20.0/service_account.yaml
kubectl apply -f https://github.com/nmstate/kubernetes-nmstate/releases/download/v0.20.0/role.yaml
kubectl apply -f https://github.com/nmstate/kubernetes-nmstate/releases/download/v0.20.0/role_binding.yaml
kubectl apply -f https://github.com/nmstate/kubernetes-nmstate/releases/download/v0.20.0/operator.yaml

Once that's done, create an NMState CR, triggering deployment of
kubernetes-nmstate handler:

cat <<EOF | kubectl create -f -
apiVersion: nmstate.io/v1beta1
kind: NMState
metadata:
  name: nmstate
EOF

v0.19.0

v0.19.0

Changes

Feature

• Add an operator that will manage the lifecycle of the existing handler. It syncs the CRDs, the namespace (if different than the operator), the RBAC, and the handler itself. (#466, @bcrochet)

Other (Bug, Cleanup or Flake)

• Adapt scc.yaml to ocp-4.5 (#502, @qinqon)

docker pull quay.io/nmstate/kubernetes-nmstate-handler:v0.19.0

v0.18.0

v0.18.0

This minor release include some bugfixes and some ports from release-0.15 like handler locking, also the
e2e tests run at workers only now.

Changes

Test

• Test: verify that bridge IP is persisted after reboot (#456, @phoracek)
• The e2e tests now run at workers and skipping master. (#483, @qinqon)
• Workflow script to run at external systems (#489, @qinqon)

Feature

• Add an exclusive lock for nmstate-handler instances (#487, @yboaron)
• New operator for kubernetes-nmstate deployment. (#461, @bcrochet)

Other (Bug, Cleanup or Flake)

• After desiredState is apply handler will wait for node to be ready in case the config change that. (#480, @qinqon)
• Bug: post rollback probes were not being executed. (#484, @qinqon)
• The nncp conditions calculations take now into account the nodes running nmstate pod instead of readiness (#477, @qinqon)

docker pull quay.io/nmstate/kubernetes-nmstate-handler:v0.18.0

v0.15.5

v0.15.5

Features:

• Add an exclusive lock for nmstate handler instances (#487) (#490)
• Adapt to external k8s (#465)
• [release-0.15] Add node ready state probe (#481)
• [release-0.15] Annotate tests with polarion IDs (#453)

Bugs:

• Ensure probes are run after success nmstatectl rollback (#485)
• test: Don't use 'sleep 5' at vlan-filtering script (#479)
• Stabilize 0.15 tests (#472)
• Pin nmstate to 0.1 (#463)

docker pull quay.io/nmstate/kubernetes-nmstate-handler:v0.15.5

v0.15.4

v0.15.4

Bugs:

• Run probe after rollback (#458)
• remove the webhook manifest (#439)

docker pull quay.io/nmstate/kubernetes-nmstate-handler:v0.15.4

v0.17.0

v0.17.0

Features:

• Add consistently check at cluster-sync (#448)
• use cluster scripts (#442)

Bugs:

• Upgrade kube-admission-webhook lib (#449), fixes race on startup

docker pull quay.io/nmstate/kubernetes-nmstate-handler:v0.17.0

v0.16.0

v0.16.0

Bump the support NetworkManager version to >= 1.22.

Features:

• Use centos8 nodes (#445)
• Use centos8 base image (#443)
• Upgrade operator sdk (#401)
• bump NetworkManager requirements to 1.22 (#418)

Bugs:

• Stabilize master tests (#446)
• Set nmstatectl 'set' timeout to (defaultGwProbe + apiServerProbe) * 2 (#444)
• Change webhook port to 54874 (#420)
• explicitly bring unavailable interfaces up (#422)
• Upgrade operator sdk (#401)
• remove user_setup (#419)
• add missing variables configuring profiling (#414)

Docs:

• bump NetworkManager requirements to 1.22 (#418)

docker pull quay.io/nmstate/kubernetes-nmstate-handler:v0.16.0

v0.15.3

v0.15.3

Patch release to increase nmstatectl set timeout with the time of the
two probes we use to check node networking is fine.

Bugs:

• Set nmstatectl 'set' timeout to (defaultGwProbe + apiServerProbe) * 2 (#434)

docker pull quay.io/nmstate/kubernetes-nmstate-handler:v0.15.3

v0.15.2

v0.15.2

This patch release disable webhook since they don't work at openshift.

Bugs:

• Disable webhooks (#431)

docker pull quay.io/nmstate/kubernetes-nmstate-handler:v0.15.2

v0.15.1

v0.15.1

Bugs:

• explicitly bring unavailable interfaces up (#423)
• Change webhook port to 54874 (#424)
• remove user_setup (#421)

docker pull quay.io/nmstate/kubernetes-nmstate-handler:v0.15.1