summaryrefslogtreecommitdiffstats
path: root/sources/hv-collector-ct
AgeCommit message (Collapse)AuthorFilesLines
2018-12-20There should be one KafkaSender per configurationPiotr Jaszczyk4-22/+46
We should keep only one instance of KafkaSender per instance. However, as the configuration might be changed (Consul update) it cannot be a strict singleton. Hence there should be 1to1 relationship beetween ConsulConfiguration and KafkaSender. Change-Id: Ie168028c4427741254b8c2fe316b82cca72d7668 Issue-ID: DCAEGEN2-1047 Signed-off-by: Piotr Jaszczyk <piotr.jaszczyk@nokia.com>
2018-12-18Add metrics for active connections countJakub Dudycz2-12/+17
* Fix and refactor gauges tests in MicrometerMetricsTests as they were not executing * Fix client disconnection handler in NettyTcpServer * Add metrics gauge and counters required to measure active connections Change-Id: I5620d398525c6859679cd5a49dc55a9fefd8b592 Signed-off-by: Jakub Dudycz <jakub.dudycz@nokia.com> Issue-ID: DCAEGEN2-1041
2018-12-18Add metric for rejected clients countFilip Krzywka2-7/+45
- renamed few counters to be more verbose about what they count - removed not needed 'total' suffix in metrics name Change-Id: I6be0201e5f39f1298706c536b12410413d49df19 Issue-ID: DCAEGEN2-1043 Signed-off-by: Filip Krzywka <filip.krzywka@nokia.com>
2018-12-17Metric: Processing timePiotr Jaszczyk4-8/+47
Add processing time metric measured as difference between "sent to DMaaP" and "WTP decoded" events. Change-Id: I73bb665145019fcca5ae36e2199ed0e1cc088fdf Issue-ID: DCAEGEN2-1036 Signed-off-by: Piotr Jaszczyk <piotr.jaszczyk@nokia.com>
2018-12-17Add metrics for dropped messagesJakub Dudycz3-29/+80
Add counters for messages dropped due to validation or undefined routing Slight refactoring Change-Id: Ibe4e38445e81babc745d7a7d95356910845293ce Signed-off-by: Jakub Dudycz <jakub.dudycz@nokia.com> Issue-ID: DCAEGEN2-1037
2018-12-14Create component tests skeleton for metricsFilip Krzywka4-9/+146
Change-Id: I5acc919171953a4bb428585a7b8420b3223aa556 Issue-ID: DCAEGEN2-1032 Signed-off-by: Filip Krzywka <filip.krzywka@nokia.com>
2018-12-14Merge "Expose Prometheus metrics"Piotr Wielebski1-1/+1
2018-12-13Expose Prometheus metricsPiotr Jaszczyk1-1/+1
Expose both JVM and HV-VES metrics by means of HealthCheckApiServer in Prometheus format. Change-Id: I81b74cd945c67569703a4a74f5a2683ede18d64a Issue-ID: DCAEGEN2-1029 Signed-off-by: Piotr Jaszczyk <piotr.jaszczyk@nokia.com>
2018-12-13Don't escape in component testsFilip Krzywka1-1/+13
We want our test output to be readable, so we should not escape \n or \t in console output. Change-Id: I4c673760ea0cd55007b47b7490036c0693887eda Issue-ID: DCAEGEN2-1026 Signed-off-by: Filip Krzywka <filip.krzywka@nokia.com>
2018-12-13Align logback configurationFilip Krzywka1-25/+42
- followed current logging guidelines in regard of format - in simulators currently very simple (minimalistic) pattern will be used as they do not contain any MDCs or Markers - in main logbacks included three patterns varying on verbosity and human-readability (for production full logging, ie. ONAP_LOG_PATTERN, should be used) - removed logback-test from domain module as it was not enabled anyway (no dependency on logback) and we should never have need for logging in this module - made logback-tests consistent (using simple pattern) except in component-tests, where it might be useful to have more verbose output (including MDCs etc.) Change-Id: Ifd851d12a77d5303b836a25c0e168e18a3a804e9 Issue-ID: DCAEGEN2-1026 Signed-off-by: Filip Krzywka <filip.krzywka@nokia.com>
2018-12-10Add log diagnostic contextPiotr Jaszczyk3-8/+9
As it's not trivial to use MDC directly from logging framework in reactive application, we need to do some work manually. The approach proposed is an explicit MDC handling, which means that context is kept as an object created after establishing client connection. Next, new instance of HvVesCollector (and its dependencies) is created. Every object is propagated with ClientContext so it can use it when calling logger methods. In the future ClientContext might be used to support other use-cases, ie. per-topic access control. As a by-product I had to refactor our Logger wrapper, too. It already had too many functions and after adding MDC number would be doubled. Change-Id: I9c5d3f5e1d1be1db66d28d292eb0e1c38d8d0ffe Issue-ID: DCAEGEN2-671 Signed-off-by: Piotr Jaszczyk <piotr.jaszczyk@nokia.com>
2018-12-05Enhance of logging in test partkjaniak1-13/+9
Change-Id: Ibdec0ac5cead7e46ada4c32983b9ccf962df703b Issue-ID: DCAEGEN2-1004 Signed-off-by: kjaniak <kornel.janiak@nokia.com>
2018-11-29Custom detekt rule for logger usage checkPiotr Jaszczyk10-0/+1045
Check if logger invocations don't use unoptimal invocations, eg. concatenation `debug("a=" + a)` instead of lambda use `debug {"a=" + a}` Unfortunately to avoid defining dependencies in many places and having circural dependencies it was necessarry to reorganize the maven module structure. The goal was to have `sources` module with production code and `build` module with build-time tooling (detekt rules among them). Issue-ID: DCAEGEN2-1002 Change-Id: I36e677b98972aaae6905d722597cbce5e863d201 Signed-off-by: Piotr Jaszczyk <piotr.jaszczyk@nokia.com>