v1.13.2

Changes

• fixes #423: Populate pod list in webhook notification #464 (thanks to Abhay Verma)
• fix(chart): added missing livenessProbe #467 (thanks to Marcus Noble)
• Add PDB and update label patterns #463 (thanks to Steve Hipwell)
• Fix Helm templates bug which caused duplicate ports key #551 (#461) (thanks to Alexander Berger)
• fix typo in helm config readme #459 (thanks to Jason Haugen)
• Upgrade deps #452 (thanks to Brandon Wagner)
• helm: Use hostPort only with hostNetwork #456 (thanks to Markos Chandras)

v1.13.1

Changes

• add pods to webhook event and log pods on node #447 (thanks to Brandon Wagner)
• feat: Enable sprig template function library to webhook template renderer #445 (thanks to Antonis Stamatiou)
• add licenses for sprig and transitive deps #446 (thanks to Brandon Wagner)
• change drop message log to WARN instead of ERR #444 (thanks to Brandon Wagner)
• make queue-processor mode replicas configurable #443 (thanks to Brandon Wagner)
• fix rolebinding kind #442 (thanks to Brandon Wagner)
• Switch IMDSv2 token warning message to debug #440 (thanks to Markos Chandras)
• fix psp role #435 (thanks to Brandon Wagner)
• always do license check and use dockerhub for go-report-card image pulls #432 (thanks to Brandon Wagner)
• remove install-awscli #431 (thanks to Bryan™)
• remove beta notice for queue-processor mode #430 (thanks to Brandon Wagner)
• remove tty on container execs #429 (thanks to Brandon Wagner)
• Populate pods list to drain event #428 (thanks to Abdul Muqtadir Mohammed)
• don't exit(1) in queue processor mode #422 (thanks to Stephen Schlie)
• Upgrade to Go 1.16 and fix report-card #427 (thanks to Bryan™)
• Correct instantiation of AWS session object #421 (thanks to Matt Dainty)
• feat: ability to emit Kubernetes events #411 (thanks to Roger Torrentsgenerós)
• Clarify EventBridge Rules setup #419 (thanks to Bryan™)
• fix: error handling for missing instances #409 (thanks to Jason Haugen)
• rename InterruptionEvent.Drained to InterruptionEvent.NodeProcessed #410 (thanks to Quan Hoang)

v1.13.0

Changes

• fixed and improved region detection within queue url #403 (thanks to Brandon Wagner)
• fix undefined log levels #402 (thanks to Brandon Wagner)
• feat: add ability to drain on rebalance #400 (thanks to Roger Torrentsgenerós)
• Switch to use SQS long polling #401 (thanks to Tristan Keen)
• SPOT_POD_IP is not used #396 (thanks to James Callahan)

v1.12.2

Changes

• Implement Liveness probe #389 (thanks to Yauheni Sliaptsou)
• Update ClusterRole to include get for pods #391 (thanks to Abdul Muqtadir Mohammed)
• Add running pods to webhook notifications #388 (thanks to Abdul Muqtadir Mohammed)
• Provide node labels for webhook template #382 (thanks to Abdul Muqtadir Mohammed)
• Provide AutoScalingGroupName for webhook notifications #384 (thanks to Abdul Muqtadir Mohammed)
• Add node name to webhook template #379 (thanks to Abdul Muqtadir Mohammed)

v1.12.1

Changes

• Drop sqs messages for absent nodes #375 (thanks to Jason Haugen)
• Update stale.yml (thanks to Jillian Kuentz)
• Add support to override pod monitor namespace #372 (thanks to Victor Boissiere)
• helm: added necessary hostPorts in the PodSecurityPolicy #371 (thanks to Maxime VISONNEAU)
• Update README.md #370 (thanks to Brandon Wagner)
• Fix goreportcard test and run make fmt #369 (thanks to Jason Haugen)
• Sqs deletion issue 360 #364 (thanks to Andrew Zenk)
• Update stale.yml (thanks to Jillian Kuentz)
• clarify queue processor install instructions #366 (thanks to Bryan™)
• Avoid noisy error on scheduled events without NotAfter #363 (thanks to Tristan Keen)
• Update stale.yml #362 (thanks to Jillian Kuentz)
• Clarify podTerminationGracePeriod docs #351 (thanks to Gabe Gorelick)
• Document that deleteLocalData defaults to true, not false #350 (thanks to Gabe Gorelick)
• add sync readme to ecr public #345 (thanks to Bryan™)
• fix release prep for gh actions #343 (thanks to Bryan™)
• Update README.md #342 (thanks to Bryan™)
• Update Chart.yaml #341 (thanks to Bryan™)

v1.12.0

New Features 🎉

• Multithreaded event processor: NTH now supports parallel processing of events when using Queue-Processor mode! The number of processors can be configured via workers parameter described here. Thanks @universam1 #317

Improvements 🚀

• Add logic to publish images to ECR on release
• Fixes to improve release process

v1.11.2

Changes

• publish images to ecr public repo (#325)

v1.11.1

Bug Fixes 🐛

• fixed volume name of daemonsets to follow DNS-1123 standard: #300 thanks @n0gu
• fixed crashes on EC2 Instance State-change Notification events: #307 thanks @universam1
• added logLevel as a helm installation parameter #312
• fixed duplicate error checking #314

v1.11.0

New Features 🎉

• Adds configurable ManagedAsgTag option allowing users to run NTH in sqs queue processing mode across multiple clusters in an account (#272 thanks @blakestoddard & @paalkr)

Improvements

• Reduce NTH startup time across all modes via lowering IMDS retry rate (#281)

v1.10.0

Introducing EC2 Instance Rebalance Recommendation 🎉🎉🎉

NTH v1.10.0 adds support for EC2 Instance Rebalance Recommendations, a notification directed towards Spot Instances that are at an elevated risk of interruption!

For v1.10.0, upon receiving the Rebalance Recommendation, NTH will cordon the node only; however, we encourage users to submit feedback if more configuration flexibility is desired!

Start consuming this new notification today with these quick-start configurations:

• IMDS Processor: --set enableRebalanceMonitoring=true
• Queue Processor: Create an Amazon EventBridge rule for EC2 Instance Rebalance Recommendations

Reference the Installation and Configuration sections of the ReadMe to learn more!

Other Changes:

• NTH Queue Processor mode now logs at debug level when a node termination event is received but the ASG is not tagged w/ the NTH managed tag aws-node-termination-handler/managed(Thanks to @kppullin #274)