Skip to content

v0.15.1
Compare
Choose a tag to compare
@github-actions github-actions released this 22 Mar 00:54
v0.15.1
2d3f351
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

0.15.1 (2024-03-22)

Features

  • add 1.27 support (c256e74)
  • add auto healing (402b23b)
  • add cluster-autoscaler (cf05ce5)
  • add image builds (2d5248b)
  • add imageRepository for container images (ccd2f6a)
  • Add labels for nodegroup name and role name (bfc2f52)
  • add m-capi-proxy (a699349)
  • add magnum certs (19cccb6)
  • Add manila csi (fceabed)
  • add tool to load image repository (56c7ca9)
  • added cluster resize (fa1e300)
  • added csi support (c9a2374)
  • added ng + create_complete (b805ad5)
  • Allow odd number of master count only (c6488c7)
  • allow using container_infra_prefix (5676b70), closes #7
  • Eliminate flux dependency for cluster autoscaler (3276d9a)
  • enable in-cluster traffic (61bf7aa)
  • Expose underlying k8s resource status to coe cluster status reason (cd0438a)
  • Fetch node count from md object when autoscaling enabled (d64ca86)
  • refactor to clusterclass (a021300)
  • RockyLinux 8/9 support (#326) (2a53f3e)
  • Support endpoint configuration for cluster-api (#211) (8a5ffac)
  • support etcd volume (#305) (544cb77)
  • Support Flatcar OS (#225) (ef4401f)
  • support keystone-auth (#297) (50a2c27)
  • Support OpenID Connect for kube-api auth (0a33863)
  • Support tls-cipher-suite configuration for kubelet (5010c9a)
  • upgrade capi version (#299) (83535e7)
  • Use crane for image loader instead of skopeo (8567f90)
  • use shorter cluster names (7b58739)
  • Validate fixed_network and fixed_subnet existence (46ac9ac)
  • Validate flavors (e438f4b)

Bug Fixes

  • Add cacert in cloud config (9fbdda6)
  • Add clusterctl installation in hack script (66f36be)
  • add containerd settings to workers (ff708dd)
  • add context to openstackmachinetemplate (6ff86b1)
  • add k8s-keystone-auth to image (0c6ca67)
  • Add missing autoscaler chart manifests (122f4b9)
  • add tests to validate rewriting manifests (c8944da)
  • add unit tests for image loader + missing images (#268) (5223b93)
  • addd 1.26.2 images (119b2c6)
  • added flux + node labels (5f04d4e)
  • added update_cluster_status (faa153b)
  • address status changes for v1alpha7 (5da7223)
  • allow cluster deletion (7dc615f)
  • allow configuring tls-cipher-suites (#261) (d1b7ab5), closes #251
  • allow for mirroring to insecure registry (3d9e364)
  • allow for optional ssh key (c2ed0af)
  • allow glance to use 10G images (f222cb1)
  • avoid overriding api_address (f7009cc)
  • build generic wheels (1f1273b)
  • bump capi (ba2866d)
  • change "fixed_subnet_cidr" default value (387fd99)
  • clean-up cluster on failures (b2f0d9e)
  • clean-up old endpoint slices (730743e)
  • cluster creation (62b89f0)
  • Convert Openstack volume type name to valid rfc1123 string (7d0f316)
  • correct 1.25 autoscaler image (52a756c)
  • correct dependency list (65e2ea1)
  • correct images (df45fc4)
  • CREATE_COMPLETE state (898c818)
  • csi: Use up-to-date provisioner name for Cinder CSI (#295) (79a0ce2)
  • deploy autoscaler right before the cluster creation completed (#307) (a48ddef)
  • disable profiling (d18936e), closes #30 #35 #36
  • Do not upgrade helmrelease in pending status (81bb06c)
  • doc: Fix k8s version string (b1f371c)
  • doc: nit picking for README (848e016)
  • Drop MachineDeployment annotation workaround #142 (e3981ed)
  • enable ssl access (7251c8a)
  • first pass at upgrades (28ce8b0)
  • fix audit log enabled clusters (#276) (ef1a1ff)
  • fix cluster status when update done (#322) (9c237d1)
  • fix jsonpatch for preKubeadmCommands (#331) (6fcf823)
  • Fix manila csi config (#254) (867efdb)
  • Fix the certificate deletion (17c243a)
  • Fix tls-insecure of ccm configuration (5c9ad68)
  • Fix typo in developer guide doc (ff59cbb)
  • helm: make upgrades more robust (d42696d)
  • helm: skip deploying autoscaler unnecessarily (6299ec5)
  • helm: upgrade path (5b9bbca)
  • image builds (434f57d)
  • incorrect case (ab16448)
  • incorrect print command (a555440)
  • mhc: increase max unhealthy to 80% (700a19c)
  • move mhc to clusterclass (c43ae93)
  • name replacement for new repo (0eaeeba)
  • only add cluster uuid to labels (76099d4)
  • only set replicas if auto scaling is disabled (a6a4bf4)
  • optimize + improve registry builds (eeb61d2)
  • pre-delete lbs (513d0ff), closes #6
  • reconcile ng status (1221e94)
  • refactor to using handlers with context (9e5cb2d)
  • relax pykube-ng requirement (10be62a)
  • remove completed todo (0d4bf03)
  • remove deleted nodegroups (66c650f)
  • remove tracebacks for missing objects (a148d4f), closes #68
  • replace the repository name (bda20a2)
  • resolve resize_cluster (7efa4f9)
  • respect availability_zone for control plane (#313) (6803bd3)
  • respect verify_ca and openstack_ca (107cc2f)
  • return helm output (95c32a7)
  • Set MachineDeployment replicas as min node count when autoscale enabled (3cc9b19)
  • Set nodegroup labels always (5d5afa5)
  • set nodeVolumeDetachTimeout property for machines (c52f3e7)
  • Set the cluster status as in_progress at the end of update_nodegroup handler (9d3312c)
  • Skip delete_cluster when stack_id is none (cabd872), closes #126
  • solve black conflict (e17ca40)
  • solve cinder-csi usage (8e9157b)
  • solve cluster delete sync (53048c8)
  • solve proxied service cleanup (476d1f7)
  • solve proxy issue with ovn (#302) (5b4bad4)
  • solve race condition for stack_id (43474f1)
  • stop adding cluster name to node name (0024683), closes #96
  • stop docker from tinkering (6fdf1d2)
  • stop unexpected rollouts (#324) (78f8a02)
  • support py3.6+ (4e1e0b5)
  • test image loader (7e7b30c)
  • Update cluster-autoscaler chart version to support k8s 1.27 (#192) (883968a)
  • update repo for initContainers (f66bb4b)
  • upgrades (bda670d)
  • use 20.04 by default (0af2de3)
  • use api from arg (386eead)
  • use apply patch strategy for kube api resource update (904abff)
  • use azure dns + fix subnet mismatch (16d75c6)
  • use capi to determine health (#226) (d599617)
  • use correct sandbox_image (80d74d2)
  • use dynamic ClusterClass version (d7fbbf0), closes #16
  • use endpoint_type for nova (688d844)
  • use getpass.getuser (67e1ec5)
  • use locking for status update (#318) (dcceec4)
  • use new registry + 1.26.2 images (b6c814f)
  • use nova_client interface (5dc34f3)
  • use operating_system (062e7f3)
  • Use public auth_url for CloudConfigSecret (4f9c852)
  • Wait until observedGeneration of the capi cluster is increased in cluster upgrade (58b6325), closes #57

Documentation

Miscellaneous Chores

Assets 2
Loading