github.com/timstclair/heapster@v0.20.0-alpha1/RELEASES.md (about) 1 # Release Notes 2 3 ## 0.18.1 (9-18-2015) 4 - Avoid using UIDs for Kubernetes events. 5 - Export container labels. 6 - Hawkular sink upgrades 7 8 ## 0.18.0 9 - Cluster Model APIs enabled by default 10 - Garbage collection in the cluster model 11 - New rolling window based bucketing storage engine for model that significantly reduces memory footprint 12 - Many fixes around race conditions with the cache 13 - GCP authorization policy updated to reduce flakiness 14 15 ## 0.17.0 16 - Fixes service account handling. 17 - New APIs that provide insight into a kubernetes cluster - usage distribution across nodes, namespaces, pods, etc. 18 19 ## 0.16.0 (7-7-2015) 20 - Add new sink for GCM autoscaling. 21 - Force heapster to use kubernetes v1.0 API. 22 - Bug fixes. 23 24 ## 0.15.0 (6-26-2015) 25 - Use service accounts while running in kubernetes clusters 26 - Expose kubernetes system containers with standard names 27 - Minor bug fixes. 28 29 ## 0.14.3 (6-19-2015) 30 - Expose HostID for nodes. 31 - Added an API '/metric-export-schema' that exposes the schema for '/metric-export' API. 32 33 ## 0.14.2 (6-16-2015) 34 - Expose HostID for all containers. 35 36 ## 0.14.1 (6-15-2015) 37 - Expose External ID of nodes as a label and namespace UID as a label. 38 - Fix bug in handling metric specific labels. 39 40 ## 0.14.0 (6-9-2015) 41 - Heapster exposes a REST endpoint that serves the metrics being pushed to sinks. 42 - Support for service accounts. 43 44 ## 0.13.0 (5-29-2015) 45 - Switch use of Kubernetes API to v1beta3. 46 - Add option to connect to Kubelets through HTTPS. 47 48 ## 0.12.1 (5-13-2015) 49 - Fixes kube master url handling via --source flag. 50 51 ## 0.12.0 (5-12-2015) 52 - Fixes issues related to supporting multiple sinks. 53 - Resource usage of kubernetes system daemons are exported. 54 - Support for kubernetes specific auth (secrets). 55 - Scalability improvements. 56 57 ## 0.11.0 (4-28-2015) 58 - Export filesystem usage metrics for the node and containers. 59 - Support for Kubernetes events 60 - New Sink - Google Cloud Logging. Supports events only. 61 - New Sink - Google Cloud Monitoring. Supports metrics only. 62 - New metric labels - 'pod_name' and 'resource_id'. 63 - Extensible source and sinks configuration. It is now possible to export metrics and events to multiple sinks simultaneously. 64 65 ## 0.10.0 (3-30-2015) 66 - Downsampling - Resolution of metrics is set to 5s by default. 67 - Support for using Kube client auth. 68 - Improved Influxdb sink - sequence numbers are generated for every metric. 69 - Reliability improvements 70 - Bug fixes. 71 72 ## 0.9 (3-13-2015) 73 - [Standardized metrics](sinks/api/supported_metrics.go) 74 - New [common API](sinks/api/types.go) in place for all external storage drivers. 75 - Simplified heapster deployment scripts. 76 - Bug fixes and misc enhancements. 77 78 ## 0.8 (2-22-2015) 79 - Avoid expecting HostIP of Pod to match node's HostIP. 80 81 ## 0.7 (2-18-2015) 82 - Support for Google Cloud Monitoring Backend 83 - Watch kubernetes api-server instead of polling for pods and nodes info. 84 - Fetch stats in parallel. 85 - Refactor code and improve testing. 86 - Miscellaneous bug fixes. 87 - Native support for CoreOS. 88 89 ## 0.6 (1-21-2015) 90 - New /validate REST endpoint to probe heapster. 91 - Heapster supports kube namespaces. 92 - Heapster uses InfluxDB service DNS name while running in kube mode. 93 94 ## 0.5 (12-11-2014) 95 - Compatiblity with updated InfluxDB service names. 96 97 ## 0.4 (12-02-2014) 98 - Compatibility with cAdvisor v0.6.2 99 100 ## 0.3 (11-26-2014) 101 - Handle updated Pod API in Kubernetes. 102 103 ## 0.2 (10-06-2014) 104 - Use kubernetes master readonly service which does not require auth 105 106 ## 0.1 (10-05-2014) 107 - First version of heapster. 108 - Native support for kubernetes and CoreOS. 109 - For Kubernetes gets pods and rootcgroup information. 110 - For CoreOS gets containers and rootcgroup information. 111 - Supports InfluxDB and bigquery. 112 - Exports pods and container stats in table 'stats' in InfluxDB 113 - rootCgroup is exported in table 'machine' in InfluxDB 114 - Special dashboard for kubernetes.