[
https://issues.apache.org/jira/browse/BEAM-7520?focusedWorklogId=437587&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-437587
]
ASF GitHub Bot logged work on BEAM-7520:
----------------------------------------
Author: ASF GitHub Bot
Created on: 27/May/20 04:10
Start Date: 27/May/20 04:10
Worklog Time Spent: 10m
Work Description: lukecwik commented on a change in pull request #9190:
URL: https://github.com/apache/beam/pull/9190#discussion_r430620228
##########
File path:
sdks/java/core/src/test/java/org/apache/beam/sdk/transforms/ParDoTest.java
##########
@@ -3488,6 +3499,158 @@ public void onTimer(OutputReceiver<String> r) {
pipeline.run();
}
+ /** A test makes sure that an event time timers are correctly ordered. */
+ @Test
+ @Category({
+ ValidatesRunner.class,
+ UsesTimersInParDo.class,
+ UsesTestStream.class,
+ UsesStatefulParDo.class,
+ UsesStrictTimerOrdering.class
+ })
+ public void testEventTimeTimerOrdering() throws Exception {
+ final int numTestElements = 100;
+ final Instant now = new Instant(1500000000000L);
+ TestStream.Builder<KV<String, String>> builder =
+ TestStream.create(KvCoder.of(StringUtf8Coder.of(),
StringUtf8Coder.of()))
+ .advanceWatermarkTo(new Instant(0));
+
+ for (int i = 0; i < numTestElements; i++) {
+ builder = builder.addElements(TimestampedValue.of(KV.of("dummy", "" +
i), now.plus(i)));
+ builder = builder.advanceWatermarkTo(now.plus(i / 10 * 10));
+ }
+
+ testEventTimeTimerOrderingWithInputPTransform(
+ now, numTestElements, builder.advanceWatermarkToInfinity());
+ }
+
+ /** A test makes sure that an event time timers are correctly ordered
using Create transform. */
+ @Test
+ @Category({
+ ValidatesRunner.class,
+ UsesTimersInParDo.class,
+ UsesStatefulParDo.class,
+ UsesStrictTimerOrdering.class
+ })
+ public void testEventTimeTimerOrderingWithCreate() throws Exception {
+ final int numTestElements = 100;
+ final Instant now = new Instant(1500000000000L);
+
+ List<TimestampedValue<KV<String, String>>> elements = new ArrayList<>();
+ for (int i = 0; i < numTestElements; i++) {
+ elements.add(TimestampedValue.of(KV.of("dummy", "" + i), now.plus(i)));
+ }
+
+ testEventTimeTimerOrderingWithInputPTransform(
Review comment:
The elements in the PCollection produced by the Create transform are
considered unordered so the stateful DoFn could see (dummy, 1) before it sees
(dummy, 0). Doesn't the DoFn need to be marked with `@RequiresTimeSortedInput`
in this case?
##########
File path:
sdks/java/core/src/test/java/org/apache/beam/sdk/transforms/ParDoTest.java
##########
@@ -3488,6 +3499,158 @@ public void onTimer(OutputReceiver<String> r) {
pipeline.run();
}
+ /** A test makes sure that an event time timers are correctly ordered. */
+ @Test
+ @Category({
+ ValidatesRunner.class,
+ UsesTimersInParDo.class,
+ UsesTestStream.class,
+ UsesStatefulParDo.class,
+ UsesStrictTimerOrdering.class
+ })
+ public void testEventTimeTimerOrdering() throws Exception {
+ final int numTestElements = 100;
+ final Instant now = new Instant(1500000000000L);
+ TestStream.Builder<KV<String, String>> builder =
+ TestStream.create(KvCoder.of(StringUtf8Coder.of(),
StringUtf8Coder.of()))
+ .advanceWatermarkTo(new Instant(0));
+
+ for (int i = 0; i < numTestElements; i++) {
+ builder = builder.addElements(TimestampedValue.of(KV.of("dummy", "" +
i), now.plus(i)));
+ builder = builder.advanceWatermarkTo(now.plus(i / 10 * 10));
+ }
+
+ testEventTimeTimerOrderingWithInputPTransform(
+ now, numTestElements, builder.advanceWatermarkToInfinity());
+ }
+
+ /** A test makes sure that an event time timers are correctly ordered
using Create transform. */
+ @Test
+ @Category({
+ ValidatesRunner.class,
+ UsesTimersInParDo.class,
+ UsesStatefulParDo.class,
+ UsesStrictTimerOrdering.class
+ })
+ public void testEventTimeTimerOrderingWithCreate() throws Exception {
+ final int numTestElements = 100;
+ final Instant now = new Instant(1500000000000L);
+
+ List<TimestampedValue<KV<String, String>>> elements = new ArrayList<>();
+ for (int i = 0; i < numTestElements; i++) {
+ elements.add(TimestampedValue.of(KV.of("dummy", "" + i), now.plus(i)));
+ }
+
+ testEventTimeTimerOrderingWithInputPTransform(
Review comment:
@je-ik @y1chi @kennknowles
The elements in the PCollection produced by the Create transform are
considered unordered so the stateful DoFn could see `(dummy, 1)` before it sees
`(dummy, 0)`. Doesn't the DoFn need to be marked with
`@RequiresTimeSortedInput` in this case?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 437587)
Time Spent: 21h 10m (was: 21h)
> DirectRunner timers are not strictly time ordered
> -------------------------------------------------
>
> Key: BEAM-7520
> URL: https://issues.apache.org/jira/browse/BEAM-7520
> Project: Beam
> Issue Type: Bug
> Components: runner-direct
> Affects Versions: 2.13.0
> Reporter: Jan Lukavský
> Assignee: Jan Lukavský
> Priority: P2
> Fix For: 2.17.0
>
> Time Spent: 21h 10m
> Remaining Estimate: 0h
>
> Let's suppose we have the following situation:
> - statful ParDo with two timers - timerA and timerB
> - timerA is set for window.maxTimestamp() + 1
> - timerB is set anywhere between <windowStart, windowEnd), let's denote that
> timerB.timestamp
> - input watermark moves to BoundedWindow.TIMESTAMP_MAX_VALUE
> Then the order of timers is as follows (correct):
> - timerB
> - timerA
> But, if timerB sets another timer (say for timerB.timestamp + 1), then the
> order of timers will be:
> - timerB (timerB.timestamp)
> - timerA (BoundedWindow.TIMESTAMP_MAX_VALUE)
> - timerB (timerB.timestamp + 1)
> Which is not ordered by timestamp. The reason for this is that when the input
> watermark update is evaluated, the WatermarkManager,extractFiredTimers() will
> produce both timerA and timerB. That would be correct, but when timerB sets
> another timer, that breaks this.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
