Hello, Dominik, Thank you for such fast response!
We tried using both the Flow Rate 1 Data Stream and data stream created with Machine Data Simulator adapter, both resulting in same error. As for the data sink, we were using the existing InfluxDB data sink. Let me know if you require any additional information. Kindest regards, Branislav ________________________________ From: Dominik Riemer <[email protected]> Sent: Friday, July 2, 2021 13:37 To: Branislav Jovicic <[email protected]>; [email protected] <[email protected]> Cc: Canteri, Michele <[email protected]>; [email protected] <[email protected]>; [email protected] <[email protected]> Subject: RE: Issue with InfluxDB Hi Branislav, thanks for reporting this! At first glance, it seems the problem comes from an error while reading the key from a nested event schema, before data is written to influx. Can you please show an example event that you are trying to persist to Influx? Then I’ll take a closer look at this issue. Are you using the standard InfluxDB sink or a custom-written Influx sink? Best, Dominik From: Branislav Jovicic <[email protected]> Sent: Friday, July 2, 2021 1:07 PM To: [email protected] Cc: Canteri, Michele <[email protected]>; [email protected]; Dominik Riemer <[email protected]>; [email protected] Subject: Issue with InfluxDB Hello, everyone, My partners (Michele and Federico, in cc) and I are trying to use an InfluxDB for one of our projects. Since we ran into an issue while trying to set everything up, we decided to try and create the test case that was presented here<https://cwiki.apache.org/confluence/display/STREAMPIPES/Description+for+E2E+Tests>, under InfluxDB section. We have configured the InfluxDB data sink according to the instructions. However, when we start the pipeline, we get the following error: pipeline-elements-all-jvm_1 | Exception in thread "Thread-7" java.lang.IllegalArgumentException: Key not found pipeline-elements-all-jvm_1 | at org.apache.streampipes.model.runtime.Event.lambda$getNestedItem$2(Event.java:97) pipeline-elements-all-jvm_1 | at org.apache.streampipes.model.runtime.Event$$Lambda$885/0x00000000541938f0.get(Unknown Source) pipeline-elements-all-jvm_1 | at java.util.Optional.orElseThrow(Optional.java:290) pipeline-elements-all-jvm_1 | at org.apache.streampipes.model.runtime.Event.getNestedItem(Event.java:97) pipeline-elements-all-jvm_1 | at org.apache.streampipes.model.runtime.Event.getFieldBySelector(Event.java:90) pipeline-elements-all-jvm_1 | at org.apache.streampipes.model.runtime.Event.getFieldBySelector(Event.java:82) pipeline-elements-all-jvm_1 | at org.apache.streampipes.sinks.databases.jvm.influxdb.InfluxDbClient.save(InfluxDbClient.java:166) pipeline-elements-all-jvm_1 | at org.apache.streampipes.sinks.databases.jvm.influxdb.InfluxDb.onEvent(InfluxDb.java:54) pipeline-elements-all-jvm_1 | at org.apache.streampipes.wrapper.standalone.runtime.StandaloneEventSinkRuntime.process(StandaloneEventSinkRuntime.java:50) pipeline-elements-all-jvm_1 | at org.apache.streampipes.wrapper.standalone.routing.StandaloneSpInputCollector.send(StandaloneSpInputCollector.java:54) pipeline-elements-all-jvm_1 | at org.apache.streampipes.wrapper.standalone.routing.StandaloneSpInputCollector.lambda$onEvent$0(StandaloneSpInputCollector.java:48) pipeline-elements-all-jvm_1 | at org.apache.streampipes.wrapper.standalone.routing.StandaloneSpInputCollector$$Lambda$881/0x0000000054170f00.accept(Unknown Source) pipeline-elements-all-jvm_1 | at java.util.concurrent.ConcurrentHashMap.forEach(ConcurrentHashMap.java:1597) pipeline-elements-all-jvm_1 | at org.apache.streampipes.wrapper.standalone.routing.StandaloneSpInputCollector.onEvent(StandaloneSpInputCollector.java:48) pipeline-elements-all-jvm_1 | at org.apache.streampipes.wrapper.standalone.routing.StandaloneSpInputCollector.onEvent(StandaloneSpInputCollector.java:29) pipeline-elements-all-jvm_1 | at org.apache.streampipes.messaging.kafka.SpKafkaConsumer.run(SpKafkaConsumer.java:120) pipeline-elements-all-jvm_1 | at java.lang.Thread.run(Thread.java:823) ... and no data gets stored in it. We checked that the data can, in fact, be stored in this DB (we manually stored some data in it) and retrieved with InfluxDB Data Stream. So, we are certain that the DB works properly, and that the issue lies in the InfluxDB data sink. In addition, on an unrelated note, we noticed that upon using the pipeline that has Notification Sink, and when we get some notifications, the notification icon still has the number of notifications displayed in it, even though we actually saw and deleted these notifications. It acts as if they were unread all the time. We believe that this is strange behavior, and we wanted to report that, as well. Let us know if you need anything else. Kindest regards, Branislav
