1.54. Release Notes 0.91

This is our first “alpha” release of NetSpyGlass. This release comes with most of the core features but is not feature complete.

We provide a single tar archive that includes the following components:

backend-1.0.jar
ui.sh
monitor.sh
nw2.conf.prototype
nw2rules.py
nw2.build
doc/Installtion_and_configuration.md
doc/config.md
doc/hbase.md
doc/logs.md
doc/monitoring_variables.md
doc/views.md
doc/tags.md
doc/licenses.txt
doc/release_notes_0.91.md

See files inside of the “doc” directory for more details on the installation and configuration of the system.

1.54.1. Main Features Available in this Release

  • see Installtion_and_configuration.md for the instructions on how to install, configure and run the program.
  • Application is configured via configuration file. See document config.md for more details on the file syntax.
  • Network discovery. The program can run discovery on schedule or on demand if user clicks a button in the UI. The program will poll only devices listed in the configuration file but will discover their interfaces, hardware components and protocols they run. It reconstructs Layer2 network topology using this information. See discovery.md for more information on the discovery schedule configuration.
  • UI: you can organize network maps into “views” that appear as tiles in application’s home page. See document views.md for more information on this.
  • Tiles in the home page display “live” network maps and are clickable. Clicking a tile opens up corresponding map.
  • hierarchical views: views can be nested (see views.md for more info). When you open a view that has nested views, the latter appear as nodes or “balls” of a different color. Clicking these opens corresponding map.
  • Each map view has breadcrumbs at the top, allowing you to navigate back to maps of the higher level.
  • once in a map, you can zoom in and out using mouse wheel, click and drag network nodes to reposition them and click anywhere in the white space between nodes and drag to drag whole map around.
  • Map legend displays available monitoring variables; changing variable in the legend causes the map to redraw itself, showing latest values for the chosen variable as network link labels.
  • Clicking network node in the map opens “Device Details” panel where you can see the table of monitoring variables for all interfaces of the device with the latest values collected by the program.
  • Clicking a number in the table opens the graph of historical values of the corresponding variable.
  • drop-down control on the left lets you choose combination of tags to filter interfaces in the table. You can build combination of several tags, which are combined using logical “OR”. Click “x” to the left of the tag word to remove it from the filter.
  • panel on the right allows you to activate predictive analysis feature. Currently we support two algorithms: linear approximation and 95-percentile. When activated, this overlays a line on top of the current graph. You can extrapolate approximation into the future as well.

1.54.2. Known Issues

  • The UI is a prototype. Although it is functional, it does not implement right UX design and lacks a lot of features.
  • All configuration is done via the configuration file
  • Even though the Monitor polls and collects data using SNMP OIDs that track parameetrs of the device as a whole or its hardware components, these do not appear on maps yet. These variables include CPU load, temperature and other things not related to network interfaces.
  • User can get graphs for the monitoring variables in the device details panel, but the graph does not provide controls to change the start time of the time interval. Time interval length can be changed though.
  • Documentation for the Data Processor component is not available yet.