v0.7.6

Bug Fixes

• use node short name to avoid unsafe truncation-then-concatenation for job-derived pod names in upstream job controller #151

v0.8.0-dev.2

job parallelism is zero unless in applying list (#145)

Attempting to address #134

Signed-off-by: Jacob Blain Christen <[email protected]>

v0.8.0-dev.1

avoid plan cache when handling jobs (#142)

Attempting to address #134

Signed-off-by: Jacob Blain Christen <[email protected]>

v0.7.5

Bug Fixes

• fix for .status.applying duplication on plans (from the fix in #126) enabling upgrades on nodes with .name != .labels["kubernetes.io/hostname"] #133

v0.7.4

Bug Fixes

• Fix for jobs failing to schedule on nodes with names that do not match the value of their kubernetes.io/hostname label #124

---

This release introduced a serious bug resulting in duplicated .status.applying entries on most plans in most clusters. Please use https://github.com/rancher/system-upgrade-controller/releases/tag/v0.7.5 instead!

v0.7.3

Merge pull request #123 from cmurphy/protobuf-bump

Bump protobuf dependency to latest

v0.7.2

Features and Enhancements

• Controller tolerations for both control-plane and controlplane #120

v0.7.1

Features and Enhancements

• Controller tolerations for control-plane and etcd node taints #117
• Controller bind mount /etc/pki (in addition to /etc/ssl) from host #116

v0.7.0

Features and Enhancements

• Bump wrangler to v0.8 #113
• Support Env and EnvFrom entries on Plans #112

Bug Fixes

• Fix for incorrect job names for nodes with FQDN #110 via #113

v0.6.2

Bug Fixes

• Prevent Job Creation for Plans that Have Not Resolved #97
• Fix Serialization of SkipWaitForDeleteTimeout Drain Option #98