Skip to content

Latest commit

 

History

History
117 lines (86 loc) · 5.43 KB

CHANGELOG.md

File metadata and controls

117 lines (86 loc) · 5.43 KB

CHANGELOG - docker_daemon

1.10.0 / 2018-05-11

  • [IMPROVEMENT] Add docker.cpu.shares metric. See [#1358][]

1.9.0 / 2018-03-23

  • [IMPROVEMENT] add the new exec_die event type to default exclusion list. See [#1240][]

1.8.0 / 2018-02-13

  • [IMPROVEMENT] Add short_image tag to container metrics. See [#986][]

1.7.0 / 2018-01-10

  • [FEATURE] Honor global collect_labels_as_tags if integration's collect_labels_as_tags is empty. See [#881][]
  • [IMPROVEMENT] Improve logging when cgroup metrics can't be retrieved. See [#914][]

1.6.0 / 2017-11-21

  • [IMPROVEMENT] Add custom tags to all service checks. See #782
  • [IMPROVEMENT] Add docker memory soft limit metric. See #760
  • [IMPROVEMENT] Add docker.containers.running.total & docker.containers.stopped.total metrics. See #859

1.5.1 / 2017-11-08

  • [BUGFIX] Fix lost kubernetes tags in 1.5.0. See #817

1.5.0 / 2017-10-10

  • [IMPROVEMENT] Remove namespace from pod_name tag. See #770

1.4.0 / 2017-09-12

  • [FEATURE] Add an option to wait for docker if it's not ready at start time. See #722
  • [FEATURE] Add client-side event filtering by event type. See #744

1.3.2 / 2017-08-28

  • [IMPROVEMENT] Add "image_name:openshift/origin-pod" to suggested exclude list. See #641
  • [BUGFIX] safely check volume list before accessing. See #544
  • [BUGFIX] make it a bit safer. See #701

1.3.1 / 2017-07-26

  • [BUGFIX] fix event collection on ecs and nomad. See #616

1.3.0 / 2017-07-18

  • [FEATURE] collect kube_container_name by default in docker_daemon check, like kubernetes does. See #553
  • [FEATURE] add kube_container_name tag to kubernetes and docker integrations. See #509, thanks @sophaskins
  • [IMPROVEMENT] remove NomadUtil & ECSUtil from docker_daemon, MetadataCollector proxies them. See #486
  • [IMPROVEMENT] use the new orchestrator.Tagger class to retrieve the Mesos tags for docker metrics. See #466

1.2.0 / 2017-06-05

  • [FEATURE] Add option to extract docker event attibutes as tags. See #404
  • [FEATURE] Add option to cap rate values to filter out cgroup CPU spikes. See #412
  • [IMPROVEMENT] Tag metrics with Nomad task/group/job names when available. See #305
  • [IMPROVEMENT] Support the new diskmapper statsd format for docker.disk.* metrics. See #409
  • [IMPROVEMENT] Report a service check when the Docker daemon is unreachable. See #354
  • [IMPROVEMENT] Add docker.cpu.usage metric for global container CPU usage. See #385
  • [BUGFIX] Docker events are tagged with the image name when docker reports its checksum. See #415

1.1.1 / 2017-05-11

  • [BUGFIX] catch IOError exception when container exits in the middle of a check run. See #408
  • [BUGFIX] fix image name when using sha256 for specs. See #393

1.1.0 / 2017-04-24

  • [FEATURE] reduce network mapping logging output. See #348
  • [FEATURE] Kubernetes: catch kubeutil init exception. See #345
  • [FEATURE] kubernetes 1.6+ cgroups hierarchy support. See #333
  • [FEATURE] ECS handle NetworkMode:host. See #320
  • [FEATURE] adds docker_network tagging to docker.net.bytes_* metrics, needs dd-agent >= 5.13.0. See #277
  • [FEATURE] adds the ability to monitor for docker exits failure with service check docker.exit. See #290
  • [FEATURE] collect rancher label container name as tag. See #282

1.0.0 / 2017-03-22

  • [FEATURE] adds docker_daemon integration.