🗞 This Week in OpenNMS - January 22nd, 2019 - Kafka Updates, WS-Man Support, UI Refresh, Helm, and more!

syslogd
dhcp
helm
minion
elasticsearch

#1

It’s time for This Week in OpenNMS!

Last week we worked on a number of Kafka-related fixes and improvements, improved minion monitor support, the Bootstrap 4 UI rework, and a bunch of other bug fixes and enhancements.

Github Project Updates

Internals, APIs, and Documentation

  • Chandra worked on an issue where syslog events could trigger notifications when they are not supposed to.
  • Antonio continued his work on exposing topology data to Kafka.
  • Jesse fixed an issue where some event parameter names would be too long to be persisted to the database.
  • Chandra did more work on supporting large buffer sizes in Kafka RPC.
  • Christian worked on modernizing the DHCP monitor to be Minion-capable.
  • Jesse continued to work on updating pushing data to Elasticsearch.
  • Chandra did more changes to the poller API support in OIA.
  • Dino worked on supporting getting provisioning asset information from WS-Man.
  • Jesse added support for printing runtime event configuration from the Karaf CLI.
  • Jesse added Helm documentation for the performance datasource.
  • Jesse fixed some issues in the Minion WS-Man support.
  • Matt continued to work on tools to help evaluate alarm and situation coverage.

Web & UI

  • Markus, Sebastian, and Ronny continued their work on moving to Bootstrap 4 for our web UI.
  • Markus fixed an issue with exceptions in OSGi-provided ReST services getting eaten.
  • Christian worked on an issue where the wallboard UI would show “Nothing to display”.
  • Markus fixed a problem where the Business Service Status panel would not show.
  • I worked on making it possible to use node or resource labels and names in Helm datasource labels.
  • Dustin did some refactoring of the code for supporting browser notifications.
  • Brynjar Eide worked on adding some missing thread pool information from Collectd and Pollerd’s instrumentation.
  • Dustin worked on making Helm’s queries to the measurements API have a consistent order.

OpenNMS Meridian and Horizon Releases

Last week we released Meridian 2018.1.4 and Horizon 23.0.2.

Both were primarily bugfix releases that also included a number of performance improvements to topology map loading.

For a complete list of changes, see the release announcements:

Until Next Week…

If there’s anything you’d like me to talk about in a future TWiO, or you just have a comment or criticism you’d like to share, don’t hesitate to say hi.

- Ben

Resolved Issues Since Last TWiO

  • HZN-1458: No stack-traces for exceptions thrown by REST services in OSGi
  • HZN-1460: Add option to syslogd to include the raw syslog message as a parameter in the generated event
  • NMS-10357: Improve Collectd and Pollerd’s ThreadPool graphs
  • NMS-10404: Support provisioning asset information via wsman
  • NMS-10441: Update favicon.ico for higher resolutions
  • NMS-10446: Support large buffer sizes in Kafka RPC
  • NMS-10449: Use the DaemonService to reload a daemon instead of the DaemonNameEnum
  • NMS-10518: refactor/clean up opennms script
  • NMS-10527: Remove unused Topology Plugin “org.opennms.features.topology.plugins.ssh”
  • NMS-10529: BusinessServiceVertexStatusInfoPanelItem is broken
  • NMS-10530: Flow query fails with NPE if it returns no data
  • OIA-1: Add support for ServiceMonitors