[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-20 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=82484=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-82484
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 20/Mar/18 21:18
Start Date: 20/Mar/18 21:18
Worklog Time Spent: 10m 
  Work Description: robertwb closed pull request #4879: [BEAM-3865] Fix 
watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879
 
 
   

This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:

As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):

diff --git 
a/examples/java/src/main/java/org/apache/beam/examples/complete/game/UserScore.java
 
b/examples/java/src/main/java/org/apache/beam/examples/complete/game/UserScore.java
index 8205baf35e3..0ebd744e0cf 100644
--- 
a/examples/java/src/main/java/org/apache/beam/examples/complete/game/UserScore.java
+++ 
b/examples/java/src/main/java/org/apache/beam/examples/complete/game/UserScore.java
@@ -128,6 +128,7 @@ public Long getTimestamp() {
 
 @ProcessElement
 public void processElement(ProcessContext c) {
+  System.out.println("GOT " + c.element());
   String[] components = c.element().split(",");
   try {
 String user = components[0].trim();
diff --git 
a/examples/java/src/main/java/org/apache/beam/examples/complete/game/injector/InjectorUtils.java
 
b/examples/java/src/main/java/org/apache/beam/examples/complete/game/injector/InjectorUtils.java
index 1667f3a4691..6bad1cbe616 100644
--- 
a/examples/java/src/main/java/org/apache/beam/examples/complete/game/injector/InjectorUtils.java
+++ 
b/examples/java/src/main/java/org/apache/beam/examples/complete/game/injector/InjectorUtils.java
@@ -86,6 +86,7 @@ public static String getFullyQualifiedTopicName(
*/
   public static void createTopic(Pubsub client, String fullTopicName)
   throws IOException {
+System.out.println("fullTopicName " + fullTopicName);
 try {
 client.projects().topics().get(fullTopicName).execute();
 } catch (GoogleJsonResponseException e) {
diff --git a/sdks/python/apache_beam/testing/data/trigger_transcripts.yaml 
b/sdks/python/apache_beam/testing/data/trigger_transcripts.yaml
index a736e943ac5..1244736c1ce 100644
--- a/sdks/python/apache_beam/testing/data/trigger_transcripts.yaml
+++ b/sdks/python/apache_beam/testing/data/trigger_transcripts.yaml
@@ -121,7 +121,7 @@ transcript:
 - input: [4]# no output
 - input: [5]
 - expect:
-- {window: [1, 14], values: [1, 2, 3, 4, 5], timestamp: 14, early: 
true}
+- {window: [1, 14], values: [1, 2, 3, 4, 5], timestamp: 15, early: 
true}
 - input: [6]
 - watermark: 100
 - expect:
diff --git a/sdks/python/apache_beam/transforms/trigger.py 
b/sdks/python/apache_beam/transforms/trigger.py
index 7d03240f941..d47c740d0e1 100644
--- a/sdks/python/apache_beam/transforms/trigger.py
+++ b/sdks/python/apache_beam/transforms/trigger.py
@@ -1083,6 +1083,7 @@ def merge(_, to_be_merged, merge_result):  # pylint: 
disable=no-self-argument
 for unused_value, timestamp in elements)
if element_output_time >= output_watermark))
   if output_time is not None:
+state.clear_state(window, self.WATERMARK_HOLD)
 state.add_state(window, self.WATERMARK_HOLD, output_time)
 
   context = state.at(window, self.clock)
diff --git a/sdks/python/apache_beam/transforms/trigger_test.py 
b/sdks/python/apache_beam/transforms/trigger_test.py
index 2247b01..a765493ae81 100644
--- a/sdks/python/apache_beam/transforms/trigger_test.py
+++ b/sdks/python/apache_beam/transforms/trigger_test.py
@@ -65,10 +65,21 @@ class TriggerTest(unittest.TestCase):
   def run_trigger_simple(self, window_fn, trigger_fn, accumulation_mode,
  timestamped_data, expected_panes, *groupings,
  **kwargs):
+# Groupings is a list of integers indicating the (uniform) size of bundles
+# to try. For example, if timestamped_data has elements [a, b, c, d, e]
+# then groupings=(5, 2) would first run the test with everything in the 
same
+# bundle, and then re-run the test with bundling [a, b], [c, d], [e].
+# A negative value will reverse the order, e.g. -2 would result in bundles
+# [e, d], [c, b], [a].  This is useful for deterministic triggers in 
testing
+# that the output is not a function of ordering or bundling.
+# If empty, defaults to bundles of size 1 in the given order.
 late_data = kwargs.pop('late_data', [])
 assert not kwargs
 
 def bundle_data(data, size):
+  if size < 0:
+data = list(data)[::-1]
+size = -size
   bundle = []
   for timestamp, elem in data:
 windows = 

[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-19 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=82017=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-82017
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 19/Mar/18 20:27
Start Date: 19/Mar/18 20:27
Worklog Time Spent: 10m 
  Work Description: charlesccychen commented on a change in pull request 
#4879: [BEAM-3865] Fix watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879#discussion_r175573879
 
 

 ##
 File path: sdks/python/apache_beam/transforms/trigger_test.py
 ##
 @@ -65,10 +65,21 @@ class TriggerTest(unittest.TestCase):
   def run_trigger_simple(self, window_fn, trigger_fn, accumulation_mode,
  timestamped_data, expected_panes, *groupings,
  **kwargs):
+# Groupings is a list of integers indicating the (uniform) size of bundles
+# to try. For example, if timestamped_data is has elements [a, b, c, d, e]
 
 Review comment:
   "is has"


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Issue Time Tracking
---

Worklog Id: (was: 82017)
Time Spent: 1h 40m  (was: 1.5h)

> Incorrect timestamp on merging window outputs.
> --
>
> Key: BEAM-3865
> URL: https://issues.apache.org/jira/browse/BEAM-3865
> Project: Beam
>  Issue Type: Bug
>  Components: sdk-py-core
>Affects Versions: 2.2.0, 2.3.0
>Reporter: Robert Bradshaw
>Assignee: Ahmet Altay
>Priority: Major
>  Time Spent: 1h 40m
>  Remaining Estimate: 0h
>
> Looks like we're setting multiple watermark holds with one arbitrarily being 
> held. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)


[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-19 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=81991=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-81991
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 19/Mar/18 19:50
Start Date: 19/Mar/18 19:50
Worklog Time Spent: 10m 
  Work Description: charlesccychen commented on a change in pull request 
#4879: [BEAM-3865] Fix watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879#discussion_r175563315
 
 

 ##
 File path: sdks/python/apache_beam/transforms/trigger_test.py
 ##
 @@ -81,16 +84,8 @@ def bundle_data(data, size):
 
 if not groupings:
   groupings = [1]
+grouping = list(groupings) + [-group_by for group_by in groupings]
 
 Review comment:
   
   
   > **robertwb** wrote:
   > Oh, yes, nice eye!
   > 
   > And fixing this makes some of the non-deterministic ones fail... Enabling 
only explicitly on some tests.
   
   
   Thanks!  Could you also add a comment at the beginning of this method on the 
behavior of `groupings`?  It is not obvious without a close reading.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Issue Time Tracking
---

Worklog Id: (was: 81991)
Time Spent: 1.5h  (was: 1h 20m)

> Incorrect timestamp on merging window outputs.
> --
>
> Key: BEAM-3865
> URL: https://issues.apache.org/jira/browse/BEAM-3865
> Project: Beam
>  Issue Type: Bug
>  Components: sdk-py-core
>Affects Versions: 2.2.0, 2.3.0
>Reporter: Robert Bradshaw
>Assignee: Ahmet Altay
>Priority: Major
>  Time Spent: 1.5h
>  Remaining Estimate: 0h
>
> Looks like we're setting multiple watermark holds with one arbitrarily being 
> held. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)


[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-19 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=81909=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-81909
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 19/Mar/18 16:09
Start Date: 19/Mar/18 16:09
Worklog Time Spent: 10m 
  Work Description: robertwb commented on a change in pull request #4879: 
[BEAM-3865] Fix watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879#discussion_r175492279
 
 

 ##
 File path: sdks/python/apache_beam/transforms/trigger_test.py
 ##
 @@ -81,16 +84,8 @@ def bundle_data(data, size):
 
 if not groupings:
   groupings = [1]
+grouping = list(groupings) + [-group_by for group_by in groupings]
 
 Review comment:
   Oh, yes, nice eye!
   
   And fixing this makes some of the non-deterministic ones fail... Enabling 
only explicitly on some tests.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Issue Time Tracking
---

Worklog Id: (was: 81909)
Time Spent: 1h 20m  (was: 1h 10m)

> Incorrect timestamp on merging window outputs.
> --
>
> Key: BEAM-3865
> URL: https://issues.apache.org/jira/browse/BEAM-3865
> Project: Beam
>  Issue Type: Bug
>  Components: sdk-py-core
>Affects Versions: 2.2.0, 2.3.0
>Reporter: Robert Bradshaw
>Assignee: Ahmet Altay
>Priority: Major
>  Time Spent: 1h 20m
>  Remaining Estimate: 0h
>
> Looks like we're setting multiple watermark holds with one arbitrarily being 
> held. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)


[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-17 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=81525=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-81525
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 17/Mar/18 07:49
Start Date: 17/Mar/18 07:49
Worklog Time Spent: 10m 
  Work Description: charlesccychen commented on a change in pull request 
#4879: [BEAM-3865] Fix watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879#discussion_r175250988
 
 

 ##
 File path: sdks/python/apache_beam/transforms/trigger_test.py
 ##
 @@ -81,16 +84,8 @@ def bundle_data(data, size):
 
 if not groupings:
   groupings = [1]
+grouping = list(groupings) + [-group_by for group_by in groupings]
 
 Review comment:
   What I mean is that the variable `grouping` (without "s" at the end) is 
never used in the rest of the function.  Is this a typo?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Issue Time Tracking
---

Worklog Id: (was: 81525)
Time Spent: 1h 10m  (was: 1h)

> Incorrect timestamp on merging window outputs.
> --
>
> Key: BEAM-3865
> URL: https://issues.apache.org/jira/browse/BEAM-3865
> Project: Beam
>  Issue Type: Bug
>  Components: sdk-py-core
>Affects Versions: 2.2.0, 2.3.0
>Reporter: Robert Bradshaw
>Assignee: Ahmet Altay
>Priority: Major
>  Time Spent: 1h 10m
>  Remaining Estimate: 0h
>
> Looks like we're setting multiple watermark holds with one arbitrarily being 
> held. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)


[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-17 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=81524=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-81524
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 17/Mar/18 07:49
Start Date: 17/Mar/18 07:49
Worklog Time Spent: 10m 
  Work Description: charlesccychen commented on a change in pull request 
#4879: [BEAM-3865] Fix watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879#discussion_r175250980
 
 

 ##
 File path: sdks/python/apache_beam/transforms/trigger_test.py
 ##
 @@ -81,16 +84,8 @@ def bundle_data(data, size):
 
 if not groupings:
   groupings = [1]
+grouping = list(groupings) + [-group_by for group_by in groupings]
 
 Review comment:
   What I mean is that the variable `grouping` (without "s" at the end) is 
never used in the rest of the function.  Is this a typo?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Issue Time Tracking
---

Worklog Id: (was: 81524)
Time Spent: 1h  (was: 50m)

> Incorrect timestamp on merging window outputs.
> --
>
> Key: BEAM-3865
> URL: https://issues.apache.org/jira/browse/BEAM-3865
> Project: Beam
>  Issue Type: Bug
>  Components: sdk-py-core
>Affects Versions: 2.2.0, 2.3.0
>Reporter: Robert Bradshaw
>Assignee: Ahmet Altay
>Priority: Major
>  Time Spent: 1h
>  Remaining Estimate: 0h
>
> Looks like we're setting multiple watermark holds with one arbitrarily being 
> held. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)


[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-17 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=81523=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-81523
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 17/Mar/18 07:49
Start Date: 17/Mar/18 07:49
Worklog Time Spent: 10m 
  Work Description: charlesccychen commented on a change in pull request 
#4879: [BEAM-3865] Fix watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879#discussion_r175250980
 
 

 ##
 File path: sdks/python/apache_beam/transforms/trigger_test.py
 ##
 @@ -81,16 +84,8 @@ def bundle_data(data, size):
 
 if not groupings:
   groupings = [1]
+grouping = list(groupings) + [-group_by for group_by in groupings]
 
 Review comment:
   What I mean is that the variable `grouping` (without "s" at the end) is 
never used in the rest of the function.  Is this a typo?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Issue Time Tracking
---

Worklog Id: (was: 81523)
Time Spent: 50m  (was: 40m)

> Incorrect timestamp on merging window outputs.
> --
>
> Key: BEAM-3865
> URL: https://issues.apache.org/jira/browse/BEAM-3865
> Project: Beam
>  Issue Type: Bug
>  Components: sdk-py-core
>Affects Versions: 2.2.0, 2.3.0
>Reporter: Robert Bradshaw
>Assignee: Ahmet Altay
>Priority: Major
>  Time Spent: 50m
>  Remaining Estimate: 0h
>
> Looks like we're setting multiple watermark holds with one arbitrarily being 
> held. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)


[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-16 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=81510=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-81510
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 17/Mar/18 05:28
Start Date: 17/Mar/18 05:28
Worklog Time Spent: 10m 
  Work Description: robertwb commented on issue #4879: [BEAM-3865] Fix 
watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879#issuecomment-373895826
 
 
   Jenkins: retest this please


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Issue Time Tracking
---

Worklog Id: (was: 81510)
Time Spent: 40m  (was: 0.5h)

> Incorrect timestamp on merging window outputs.
> --
>
> Key: BEAM-3865
> URL: https://issues.apache.org/jira/browse/BEAM-3865
> Project: Beam
>  Issue Type: Bug
>  Components: sdk-py-core
>Affects Versions: 2.2.0, 2.3.0
>Reporter: Robert Bradshaw
>Assignee: Ahmet Altay
>Priority: Major
>  Time Spent: 40m
>  Remaining Estimate: 0h
>
> Looks like we're setting multiple watermark holds with one arbitrarily being 
> held. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)


[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-16 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=81509=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-81509
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 17/Mar/18 05:27
Start Date: 17/Mar/18 05:27
Worklog Time Spent: 10m 
  Work Description: robertwb commented on a change in pull request #4879: 
[BEAM-3865] Fix watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879#discussion_r175248717
 
 

 ##
 File path: sdks/python/apache_beam/transforms/trigger_test.py
 ##
 @@ -81,16 +84,8 @@ def bundle_data(data, size):
 
 if not groupings:
   groupings = [1]
+grouping = list(groupings) + [-group_by for group_by in groupings]
 
 Review comment:
   It expands grouping by also adding negative values, which also tests adding 
elements in reverse order. 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Issue Time Tracking
---

Worklog Id: (was: 81509)
Time Spent: 0.5h  (was: 20m)

> Incorrect timestamp on merging window outputs.
> --
>
> Key: BEAM-3865
> URL: https://issues.apache.org/jira/browse/BEAM-3865
> Project: Beam
>  Issue Type: Bug
>  Components: sdk-py-core
>Affects Versions: 2.2.0, 2.3.0
>Reporter: Robert Bradshaw
>Assignee: Ahmet Altay
>Priority: Major
>  Time Spent: 0.5h
>  Remaining Estimate: 0h
>
> Looks like we're setting multiple watermark holds with one arbitrarily being 
> held. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)


[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-16 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=81435=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-81435
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 16/Mar/18 23:37
Start Date: 16/Mar/18 23:37
Worklog Time Spent: 10m 
  Work Description: charlesccychen commented on a change in pull request 
#4879: [BEAM-3865] Fix watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879#discussion_r175236867
 
 

 ##
 File path: sdks/python/apache_beam/transforms/trigger_test.py
 ##
 @@ -81,16 +84,8 @@ def bundle_data(data, size):
 
 if not groupings:
   groupings = [1]
+grouping = list(groupings) + [-group_by for group_by in groupings]
 
 Review comment:
   
   What is this?  This is currently a no-op.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Issue Time Tracking
---

Worklog Id: (was: 81435)
Time Spent: 20m  (was: 10m)

> Incorrect timestamp on merging window outputs.
> --
>
> Key: BEAM-3865
> URL: https://issues.apache.org/jira/browse/BEAM-3865
> Project: Beam
>  Issue Type: Bug
>  Components: sdk-py-core
>Affects Versions: 2.2.0, 2.3.0
>Reporter: Robert Bradshaw
>Assignee: Ahmet Altay
>Priority: Major
>  Time Spent: 20m
>  Remaining Estimate: 0h
>
> Looks like we're setting multiple watermark holds with one arbitrarily being 
> held. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)


[jira] [Work logged] (BEAM-3865) Incorrect timestamp on merging window outputs.

2018-03-16 Thread ASF GitHub Bot (JIRA)

 [ 
https://issues.apache.org/jira/browse/BEAM-3865?focusedWorklogId=81341=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-81341
 ]

ASF GitHub Bot logged work on BEAM-3865:


Author: ASF GitHub Bot
Created on: 16/Mar/18 20:12
Start Date: 16/Mar/18 20:12
Worklog Time Spent: 10m 
  Work Description: robertwb commented on issue #4879: [BEAM-3865] Fix 
watermark hold handling bug.
URL: https://github.com/apache/beam/pull/4879#issuecomment-373831533
 
 
   R: @charlesccychen 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Issue Time Tracking
---

Worklog Id: (was: 81341)
Time Spent: 10m
Remaining Estimate: 0h

> Incorrect timestamp on merging window outputs.
> --
>
> Key: BEAM-3865
> URL: https://issues.apache.org/jira/browse/BEAM-3865
> Project: Beam
>  Issue Type: Bug
>  Components: sdk-py-core
>Affects Versions: 2.2.0, 2.3.0
>Reporter: Robert Bradshaw
>Assignee: Ahmet Altay
>Priority: Major
>  Time Spent: 10m
>  Remaining Estimate: 0h
>
> Looks like we're setting multiple watermark holds with one arbitrarily being 
> held. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)