GitLab

- When building the SensorNavigator, trailing whitespace must be
stripped from the sensor names if they were shorter than 16 characters

- Fixed a bug that prevented pipelines in collectagent deployments
- Incremented the tolerance threshold of fuzzy queries to 1h
- Fixed the assignment of MQTT topics to job sensors

- Fixed a bug that prevented relaxed mode from working properly for
operators deployed on the collectagent
- Changed the function of the "delay" parameter: it now specifies a
fixed offset applied to the computation interval of each operator,
useful to tune the flow of data in operator pipelines

- Now using references instead of pointers
- Return type is now bool

- If users specify a single timestamp (startTs==endTs) when performing
a sensor query through dcdbquery or libdcdb, a fuzzy sensor query is
performed, and the closest sensor reading to the target timestamp is
returned

- A regular expression (jobFilter) can be applied to the jobs' nodelists
to identify the jobs that need to be treated by each collectagent in a
cluster
- By default only the first node in the nodelist is considered, but this
behavior can be customized

- querySensor does not return NULL pointers anymore, but always
the "buffer" input argument
- the previous implementation was more descriptive of error statuses,
but much more prone to leaking memory in case of improper usage
- Also fixed a warning in the MSR plugin

- Fixed a bug in the collectagent that caused the QueryEngine to not
be treated correctly, due to a missing linking flag

- Make clean might be required

- Now it is possible to perform queries on raw topics without doing
a lookup on the publishedsensors table, by using the -d option

- Fixed a fair amount of bugs related to job analyzers
- Added a "getJobsInIntervalRunning" method to JobDataStore, to get
the list of jobs that were running in a certain time interval
- Some code refactoring for the JobDataStore class to reduce duplication
- Job analyzers have been tested and are working properly as of now

- Removed the dlclose call on plugin unloads to avoid segmentation fault
issues at destruction time of boost thread objects due to failed symbol resolution
- QueryEngine in DCDBPusher does not block anymore while being updated,
but simply returns NULL

- Now supports also standard deviation and percentile computation
- Multiple output sensors per analyzer can be defined, each performing
a different aggregation task
- QueryEngine has been slightly modified so as not to clear the input
sensor reading buffer (if any), thus allowing to accumulate sensor
values with subsequent calls

- In the new implementation the sensor map used to speed up access to
sensor data is rebuilt synchronously when plugins are reloaded/loaded/unloaded
- Any threads accessing the sensor query callback will block until
the update has finished

- New method is lock-based and compatible with all features (e.g. REST API)
in the framework
- Performance impact is expected to be negligible as race conditions will
not be frequent

- Template for Job Analyzers, derived from AnalyzerInterface

Move CXXFLAGS definitions to config.mk and thus streamline compiler flags for all Makefiles in subdirectories

Remove  maxConnPerHost and maxConcRequests configuration parameters in Collect Agent as they have been deprecated in Cassandra

- This feature is not needed anymore in the new REST API design, and
forward slashes are parsed correctly without additional logic

- This feature is not needed anymore in the new REST API design, and
forward slashes are parsed correctly without additional logic

- Analytics plugins can be now loaded and unloaded dynamically
- DCDBPusher now looks into the current configuration directory if no
path is supplied when loading/reloading a plugin

- Changes mostly affect the REST API

- Added documentation to the new SensorID implementation
- NOTE: virtual sensors will likely require some modifications due to
how topics are handled now ('/' separator clashing with divide symbol)

- String topics are now used across all of DCDB, up to Cassandra
- As of now, sensor names ALWAYS match MQTT topics
- Users can still define sensor "aliases" via dcdbconfig that point
to the same topics

- Update config files to reflect new topic structure
- Updated READMEs accordingly
- Added support for "relaxed" analyzers, whose units can be built
even if some sensors cannot be found in the system
- Added support for sub-units, so as to build hierarchical unit structures