Changelog

Release 3.16.0 (12 December 2017)

  • Frontend
    • Deserializing of dashboards is now async, as a result dashboards load smoother
    • Improvements to the creation of widgets UI
      • Dimensions are now sorted by name and the values are sorted in the dropdown
      • Searching now shows matches in other data sources
      • Metric groups show above metrics
      • The height of the dimension selector is now responsive
      • Short form of the server pickers was improved
    • Dashboard will retry loading if initial loading fails
  • Agent & Plugins
    • A user can now delete an agent with the servers & containers linked to it
    • Resource plugin
      • Get filesystem metrics without having to mount / inside the container
  • Streaming backend & datastore

Release 3.15.0 (23 November 2017)

  • Frontend
    • New visualizations to for resource quotas for your containers in Kubernetes and OpenShift
    • Added a widget to show detailed information about the container on the container detail dashboard
    • Users should now be able to add labels to multiple applications
    • Servers page now lazy loads the servers
    • Windows agent install instructions now contain a download link
  • API
    • API should have less issues when big container clusters connect for the first time
    • Fixed a bug where servers would sometimes not get the HTTP_SERVER attribute
  • Agent & Plugins
    • New Plugin: .NET Core
      • Allows you to gather detailed metrics about your containers running .NET
    • Docker
      • Plugin now handles servers being turned off and CRON worker will clean up after agents have been turned off
      • New metric: Number of images
    • Kubernetes
      • Namespaces are now in their own server group. Allowing you to differentiate between namespaces and Nodes better
    • Cassandra
      • 2 new metrics added (Read requests per second, Write requests per second) and the types of multiple metrics has changed
    • Log plugin
      • A wildcard can now be used for the location of the log file
  • Code stability improvements
    • Upgrade to python3.5
    • New version of ebean
    • All javascript is now compiled into 1 file
    • Upgraded to thirft 0.10
  • On Premise
    • Removed the warning from the Cassandra container about threshold for batch insertions

Release 3.14.0 (17 October 2017)

  • Frontend
    • Uptime widget: new widget to see the uptime in a clear overview. Works with existing metrics and the new metrics from the HTTP endpoint plugin
    • Servers set to monitoring OFF are no longer shown in the topology
    • Dashboard loading optimizations
    • Possiblity to set a support email
  • Agent plugins
    • New plugins
      • Dropwizard: added support for monitoring the Dropwizard Java framework
      • HTTP endpoint: a plugin to monitor different HTTP endpoints with our agent
      • Hana DB: new metrics to monitor Hana DB
    • Kubernetes plugin:
      • New metric: number of pods created in namespace, in a period of time
      • Event for Kubelet service restarts
      • Hostname of the Kubernetes master will now be “kubernetes-master” in COSCALE_KUBE_MON_MASTER environments
    • Resources plugin
      • New metric: number of zombie processes
      • New metrics for network interfaces: network interface up/down, network interface has ipv4, network interface has ipv6, packet loss count
  • On premise
    • Initial support for easy migrations
    • Sensible memory limits on all Java processes

Release 3.13.0 (25 September 2017)

  • Kibana integration
    • You can now interactively link your Kibana instance with CoScale
  • Agent plugins
    • New plugin: Calico
    • Jolokia plugin:
      • Added native support for Openshift Jolokia
    • Kubernetes plugin:
      • Time in state metrics converted to an event, significant performance improvements as a result.
    • Docker plugin:
      • New metric: Total number of containers
      • Docker plugin publically available for Microsoft Windows 2016
  • Frontend
    • Better warning and error messages when using docker labels
    • When disconnecting a dimension from the global selector your selection is no longer lost
    • Servers page: Page should load a lot faster now
  • CoScale CLI now available for Mac OS X

Release 3.12.0 (12 September 2017)

  • Alert page improvements
    • Many improvements including: Clone alert triggers, create dashboard from alert
  • UX improvements & Small css improvements
    • Consistency in the plugin dialogs
    • Improvements to the flow for SSO users
    • Users can change their application name
    • Agent page: Always show configure and install instructions buttons
    • Widgets / Topology: Show the server the container is running on in the tooltip
  • Agents
    • New plugins: Apache Solr, Jolokia plugin
    • Optimize sending data when agent has been offline for a while
    • Resources plugin: added more metrics
    • JMX plugin
      • Ability to add multiple endpoints
      • Support authentication
    • Docker/Kubernetes plugin
      • Too few replica events for DaemonSets and StatefulSets
      • CoScale Docker images are now certified for Red Hat
      • “Docker total host cpu usage” metrics added
      • Change ‘groups’ in security context constraint so that only ‘system:cluster-admins’ are allowed to use it.
      • Moved POD system containers to another server group

Release 3.11.0 (30 June 2017)

  • Frontend improvements
    • Allow enterprise users to login to CoScale with their Single-Sign-On (SSO) platform
    • [Public dashboard] Add subtile CoScale branding on public dashboard
    • Improve user experience for selecting metrics and events in the wizard.
    • Added support for COUNT/COUNTER metrics
  • Agent improvements
    • Monitor internal kubernetes services (etcd, kubeapi)
    • [Postgresql] Add support to gather query performance data from pg_stats
  • Anomaly detection
    • Improve anomaly detection for thread based metrics
    • Reduce sensitivity for RUM

Release 3.10.0 (17 May 2017)

  • Alerts through webhooks.
    • Allows integration with many new alert providers (PagerDuty, OpsGenie, Nagios)
  • Agent improvements
    • Docker image detection now supports more flexible regex’s
    • Improved UI to provide detected images during the configuration
    • Docker plugin startup and image detection got some significant performance improvements
    • The agent now skips duplicate UPDATE_CONFIG agent actions
    • Improved logging in the agent, now includes containerid for easier debugging
  • Alerts
    • Default alerts for CPU is now average over all cores, not for separate cores
  • Switched over from Intercom to ZenDesk to provide support
  • OnPrem
    • Update script (apicall) now correctly resets the api configuration after completion.
    • Pre-allocator no longer runs on 2 containers preventing duplicate column family ID’s in cassandra

Release 3.9.0 (25 April 2017)

  • New default dashboards for the Docker & container orchestrators
    • Technology pane now has a click through
    • Topology widget now has a click through
    • Default dashboards for Docker added
  • Improvements in the Agents
    • Agent now gathers more detailed metrics about deployments in Kubernetes & OpenShift environments
    • Plugin configurations can be added to your Docker images as labels and will now be picked up by the agent
  • Improvements to the anomaly detection
    • Widgets on the dashboards will now show related anomalies
    • Anomaly emails are now more informative and can be further explored with a link to the app
    • Loadbalanced anomalies now also work on latencies
  • General improvements to the frontend
    • Dashboard topbar was redesigned to be more responsive and have reduced clutter
    • EventList is now has a search field and received many small style improvements
    • List of dashboards is now autorefreshed.
  • Performance improvements to speed up the api
    • Alerts should no longer trigger on disabled servers
  • Alerts can now be set on the messages of events
  • Billing page now allows for a registration key to be set

Release 3.8.0 (10 March 2017)

  • Inline configuration on the agents page & new configuration for custom agents
  • Anomalies show on all widgets
  • New collector: Azure

Release 3.7.0 (27 February 2017)

  • Using Kubernetes Watcher api to reduce Kubernetes plugin overhead
  • Collect container metrics on hosts instead of Kubernetes api to improve scalability
  • Support for ReplicaSets and StatefulSets
  • Various fixes to plugins and bar charts

Release 3.6.0 (20 February 2017)

  • New and improved flow to create widgets
  • HTTPS support for on-prem installations

Release 3.5.0 (23 January 2016)

  • A new plugin for Couchbase
  • Monitoring /dev/stdout and /dev/stderr from Docker plugin
  • Monitoring Kubernetes Cron jobs

Release 3.4.0 (12 January 2016)

  • New plugin: ActiveMQ
  • General usability improvements throughout the frontend
    • Dashboard dropdown can now be navigated via keyboard navigation.
    • Usability improvements for the event list widget
    • Improved error pages
  • Alerts email now support dimensions
  • OpenShift plugin now has new authentication options
  • Zooming on charts should no longer break

Release 3.3.0 (15 December 2016)

  • A user-friendly agent installation flow for docker environments
  • Dimension selector for Requests and RUM pages
  • Custom visualisation for anomalies on load-balanced services
  • Alerting page that allows the user to resolve multiple alerts at once
  • Separate plugin for OpenShift
  • Improved functionality on JMX plugin
  • Lots of small improvements and bugfixes

Release 3.2.0 (23 November 2016)

  • Active checks in all our plugins: this allows you to check whether a service is up and how fast it is responding to your health check request
  • Improved docker monitoring by introducing a table view on the containers that are consuming the most resources
  • New metrics to find containers that are hanging in a pending or creating state
  • Support for Google Container Engine
  • Support for multiple memcached services on 1 server
  • Lots of small improvements and bugfixes

Release 3.1.0 (20 October 2016)

  • Improved the dynamic behaviour of server and server groups : we leverage the start and stop times on the containers to keep the number of containers fetched and displayed by the frontend as low as possible. We also added a start and stop time to server group, so that for example old Kubernetes services don’t stay around forever,
  • Added support for multiple Redis servers in 1 agent,
  • Further improved the resource usage of the CoScale Agent.
  • Lots of small improvements and bugfixes