[jira] [Commented] (SPARK-24699) Watermark / Append mode should work with Trigger.Once
[ https://issues.apache.org/jira/browse/SPARK-24699?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16539784#comment-16539784 ] Apache Spark commented on SPARK-24699: -- User 'tdas' has created a pull request for this issue: https://github.com/apache/spark/pull/21746 > Watermark / Append mode should work with Trigger.Once > - > > Key: SPARK-24699 > URL: https://issues.apache.org/jira/browse/SPARK-24699 > Project: Spark > Issue Type: Bug > Components: Structured Streaming >Affects Versions: 2.3.1 >Reporter: Chris Horn >Priority: Major > Attachments: watermark-once.scala, watermark-stream.scala > > > I have a use case where I would like to trigger a structured streaming job > from an external scheduler (once every 15 minutes or so) and have it write > window aggregates to Kafka. > I am able to get my code to work when running with `Trigger.ProcessingTime` > but when I switch to `Trigger.Once` the watermarking feature of structured > streams does not persist to (or is not recollected from) the checkpoint state. > This causes the stream to never generate output because the watermark is > perpetually stuck at `1970-01-01T00:00:00Z`. > I have created a failing test case in the `EventTimeWatermarkSuite`, I will > create a [WIP] pull request on github and link it here. > > It seems that even if it generated the watermark, and given the current > streaming behavior, I would have to trigger the job twice to generate any > output. > > The microbatcher only calculates the watermark off of the previous batch's > input and emits new aggs based off of that timestamp. > This state is not available to a newly started `MicroBatchExecution` stream. > Would it be an appropriate strategy to create a new checkpoint file with the > most up to watermark or watermark and query stats? -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-24699) Watermark / Append mode should work with Trigger.Once
[ https://issues.apache.org/jira/browse/SPARK-24699?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16528356#comment-16528356 ] Chris Horn commented on SPARK-24699: I have attached two scala repl scripts for reproducing this behavior. The "once" variant fails to produce output or updated watermarks, the "stream" variant behaves mostly as expected. > Watermark / Append mode should work with Trigger.Once > - > > Key: SPARK-24699 > URL: https://issues.apache.org/jira/browse/SPARK-24699 > Project: Spark > Issue Type: Bug > Components: Structured Streaming >Affects Versions: 2.3.1 >Reporter: Chris Horn >Priority: Major > Attachments: watermark-once.scala, watermark-stream.scala > > > I have a use case where I would like to trigger a structured streaming job > from an external scheduler (once every 15 minutes or so) and have it write > window aggregates to Kafka. > I am able to get my code to work when running with `Trigger.ProcessingTime` > but when I switch to `Trigger.Once` the watermarking feature of structured > streams does not persist to (or is not recollected from) the checkpoint state. > This causes the stream to never generate output because the watermark is > perpetually stuck at `1970-01-01T00:00:00Z`. > I have created a failing test case in the `EventTimeWatermarkSuite`, I will > create a [WIP] pull request on github and link it here. > > It seems that even if it generated the watermark, and given the current > streaming behavior, I would have to trigger the job twice to generate any > output. > It seems like the microbatcher only calculates the watermark off of the > previous batch's input and emits new aggs based off of that timestamp. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-24699) Watermark / Append mode should work with Trigger.Once
[ https://issues.apache.org/jira/browse/SPARK-24699?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16528355#comment-16528355 ] Apache Spark commented on SPARK-24699: -- User 'c-horn' has created a pull request for this issue: https://github.com/apache/spark/pull/21676 > Watermark / Append mode should work with Trigger.Once > - > > Key: SPARK-24699 > URL: https://issues.apache.org/jira/browse/SPARK-24699 > Project: Spark > Issue Type: Bug > Components: Structured Streaming >Affects Versions: 2.3.1 >Reporter: Chris Horn >Priority: Major > Attachments: watermark-once.scala, watermark-stream.scala > > > I have a use case where I would like to trigger a structured streaming job > from an external scheduler (once every 15 minutes or so) and have it write > window aggregates to Kafka. > I am able to get my code to work when running with `Trigger.ProcessingTime` > but when I switch to `Trigger.Once` the watermarking feature of structured > streams does not persist to (or is not recollected from) the checkpoint state. > This causes the stream to never generate output because the watermark is > perpetually stuck at `1970-01-01T00:00:00Z`. > I have created a failing test case in the `EventTimeWatermarkSuite`, I will > create a [WIP] pull request on github and link it here. > > It seems that even if it generated the watermark, and given the current > streaming behavior, I would have to trigger the job twice to generate any > output. > It seems like the microbatcher only calculates the watermark off of the > previous batch's input and emits new aggs based off of that timestamp. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
