v0.13.0

[autoscaler]

🏃 Others

• [USER] Support scale to/from zero for MCM OOT providers - AWS, Azure. (#65, @prashanth26)
• [OPERATOR] Ignore gardener-specific and csi-specific labels while comparing nodegroups (#62, @hardikdr)

v0.12.0

[autoscaler]

Improvements

• [OPERATOR] MCM provider now also returns unregistered nodes to Autoscaler. This change enables autoscaler to pick up an alternate worker-pool if the chosen one can't be scaled-up. (#58, @jsravn)
• [OPERATOR] Autoscaler avoids scaling down the machines which are already being terminated. (#59, @hardikdr)

v0.11.0

[autoscaler]

Improvements

• [OPERATOR] Rebased cluster autoscaler with upstream and the default branch is now machine-controller-manager-provider.

v0.10.0

[autoscaler]

Improvements

• [OPERATOR] Add topology.kubernetes.io labels to be ignored when comparing similar node groups. (#50, @hardikdr)
• [OPERATOR] Prepended v in the version. (#42, @hardikdr)

cluster-autoscaler-chart-1.0.2

Scales Kubernetes worker nodes within autoscaling groups.

0.9.0

[autoscaler]

Improvements

• [OPERATOR] Avoid checking controller references while draining the node. (#44, @hardikdr)

0.8.0

[autoscaler]

Improvements

• [USER] Add configurable delay for pod age before considering for scale-up (#33, @hardikdr)
• [OPERATOR] Update azure instance types in cluster-autoscaler. (#41, @hardikdr)
• [OPERATOR] Updated the AWS-instance types used cluster autoscaler. (#40, @hardikdr)
• [OPERATOR] Cluster Autoscaler ignores the worker pools with maxSize set to 0. (#39, @hardikdr)
• [OPERATOR] Autoscaler uses zone-information while scaling-up the node-group from zero (#38, @hardikdr)

0.7.1

[autoscaler]

Improvements

• [OPERATOR] Avoid checking controller references while draining nodes (#45, @prashanth26)

0.7.0

[autoscaler]

Improvements

• [USER] Cluster Autoscaler now supports scale to/from 0 for MCM provider (applicable to only AWS & Azure for now) (#26, @rewiko)
• [USER] Cluster Autoscaler skips the nodes not managed by MCM. (#25 , @rewiko)

0.6.0

[autoscaler]

Improvements

• [OPERATOR] Ignore worker specific labels and 128Ki of memory difference while balancing nodegroups. (#23, @hardikdr)