[
https://issues.apache.org/jira/browse/BEAM-13025?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17442448#comment-17442448
]
Brian Hulette commented on BEAM-13025:
--------------------------------------
I looked at the most recent failing job. It has the errors that [~apilloud]
quoted. They must be red herrings though. It also has this:
{code}
generic::unknown: org.apache.beam.sdk.util.UserCodeException:
java.lang.IllegalArgumentException: Attempted to set an event-time timer with
an output timestamp of 294247-01-09T04:00:54.775Z that is after the timer
firing timestamp 2021-11-12T18:55:12.516Z
at
org.apache.beam.sdk.util.UserCodeException.wrap(UserCodeException.java:39)
at
org.apache.beam.sdk.transforms.Deduplicate$DeduplicateFn$DoFnInvoker.invokeProcessElement(Unknown
Source)
at
org.apache.beam.fn.harness.FnApiDoFnRunner.processElementForWindowObservingParDo(FnApiDoFnRunner.java:771)
at
org.apache.beam.fn.harness.data.PCollectionConsumerRegistry$MetricTrackingFnDataReceiver.accept(PCollectionConsumerRegistry.java:257)
at
org.apache.beam.fn.harness.data.PCollectionConsumerRegistry$MetricTrackingFnDataReceiver.accept(PCollectionConsumerRegistry.java:209)
at
org.apache.beam.fn.harness.BeamFnDataReadRunner.forwardElementToConsumer(BeamFnDataReadRunner.java:172)
at
org.apache.beam.sdk.fn.data.BeamFnDataInboundObserver2.awaitCompletion(BeamFnDataInboundObserver2.java:126)
at
org.apache.beam.fn.harness.control.ProcessBundleHandler.processBundle(ProcessBundleHandler.java:467)
at
org.apache.beam.fn.harness.control.BeamFnControlClient.delegateOnInstructionRequestType(BeamFnControlClient.java:151)
at
org.apache.beam.fn.harness.control.BeamFnControlClient$InboundObserver.lambda$onNext$0(BeamFnControlClient.java:116)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
Caused by: java.lang.IllegalArgumentException: Attempted to set an event-time
timer with an output timestamp of 294247-01-09T04:00:54.775Z that is after the
timer firing timestamp 2021-11-12T18:55:12.516Z
at
org.apache.beam.vendor.guava.v26_0_jre.com.google.common.base.Preconditions.checkArgument(Preconditions.java:440)
at
org.apache.beam.fn.harness.FnApiDoFnRunner$FnApiTimer.getTimerForTime(FnApiDoFnRunner.java:1914)
at
org.apache.beam.fn.harness.FnApiDoFnRunner$FnApiTimer.setRelative(FnApiDoFnRunner.java:1839)
at
org.apache.beam.sdk.transforms.Deduplicate$DeduplicateFn.processElement(Deduplicate.java:318)
{code}
It looks like there's an issue with Deduplicate on Runner v2. There was a
suspicious change to Deduplicate recently too:
https://github.com/apache/beam/commit/ce3a5545e1ac5a655a2c01374b89c08bf5b3e34a#diff-6a2e50eb57656ea50a5faa1a0346af656bee517103c3320e0ad08d6cdb2778b5
[~dpcollins-google] will see if we can run the test successfully without
Deduplicate
> beam_PostCommit_Java_DataflowV2 failing pubsublite.ReadWriteIT
> --------------------------------------------------------------
>
> Key: BEAM-13025
> URL: https://issues.apache.org/jira/browse/BEAM-13025
> Project: Beam
> Issue Type: Bug
> Components: test-failures
> Reporter: Kyle Weaver
> Assignee: Daniel Collins
> Priority: P1
> Labels: currently-failing
>
> [https://ci-beam.apache.org/job/beam_PostCommit_Java_DataflowV2/758/testReport/org.apache.beam.sdk.io.gcp.pubsublite/ReadWriteIT/testReadWrite/]
> java.lang.AssertionError: Did not receive signal on
> projects/apache-beam-testing/subscriptions/result-subscription--5335365384640437489
> in 300s
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
