Hello,
I created a PR #6672 (https://github.com/apache/flink/pull/6672)
for FLINK-9523.
Could anyone take a look at it please?
Thank you!
--
best regards,
Mikhail Sokolov
Piotr Nowojski created FLINK-10837:
--
Summary: Kafka 2.0 test KafkaITCase.testOneToOneSources dead
locked on deleteTestTopic?
Key: FLINK-10837
URL: https://issues.apache.org/jira/browse/FLINK-10837
Pr
Piotr Nowojski created FLINK-10838:
--
Summary: Rewrite Kafka tests that fail Kafka brokers
Key: FLINK-10838
URL: https://issues.apache.org/jira/browse/FLINK-10838
Project: Flink
Issue Type: I
Hi Jincheng,
Thanks for the summary!
I like the approach with append() better than the implicit forwarding as it
clearly indicates which fields are forwarded.
However, I don't see much benefit over the flatMap(Expression*) variant, as
we would still need to analyze the full expression tree to ensu
Thanks for letting me know Jincheng. I tried to reproduce the problem but I
was so far not successful. I will continue the investigation. For the
moment I would not block the release on that.
Cheers,
Till
On Thu, Nov 8, 2018 at 7:19 PM jincheng sun
wrote:
> Hi Till,
>
> Today when I do the CI b
Stefan Richter created FLINK-10839:
--
Summary: Fix implementation of PojoSerializer.duplicate() w.r.t.
subclass serializer
Key: FLINK-10839
URL: https://issues.apache.org/jira/browse/FLINK-10839
Proje
vinoyang created FLINK-10840:
Summary: BucketingSink incorrectly clears the pendingFiles List
Key: FLINK-10840
URL: https://issues.apache.org/jira/browse/FLINK-10840
Project: Flink
Issue Type: Bu
Pawel Bartoszek created FLINK-10841:
---
Summary: Reduce the number of ListObjects calls when checkpointing
to S3
Key: FLINK-10841
URL: https://issues.apache.org/jira/browse/FLINK-10841
Project: Flink
Andrey Zagrebin created FLINK-10842:
---
Summary: Waiting loops are broken in e2e/common.sh
Key: FLINK-10842
URL: https://issues.apache.org/jira/browse/FLINK-10842
Project: Flink
Issue Type: B
Timo Walther created FLINK-10843:
Summary: Make Kafka version definition more flexible for new Kafka
table factory
Key: FLINK-10843
URL: https://issues.apache.org/jira/browse/FLINK-10843
Project: Flin
Yes, Till, I am sorry that I did not keep the exception information before
re-run the CI. When I started CI again, the problem has disappeared. I will
continue to pay attention to this issue. I will keep the log when it
happens again.
I agree that this issue is not the release blocker !
Best,
Jin
Daniel Harper created FLINK-10844:
-
Summary: ArrayIndexOutOfBoundsException during checkpoint (Flink
1.5.2)
Key: FLINK-10844
URL: https://issues.apache.org/jira/browse/FLINK-10844
Project: Flink
Hi,
Yes, it seems like the best solution. Maybe someone else can also suggests if
we can split it further? Maybe changes in the interface in one doc, reading
from hive meta store another and final storing our meta informations in hive
meta store?
Piotrek
> On 9 Nov 2018, at 01:44, Zhang, Xuef
Hi
Good point with select/epoll, however I do not see how they couldn’t be with
Flink if we would like single task in Flink to be single-threaded (and I
believe we should pursue this goal). If your connector blocks on `select`, then
it can not process/handle control messages from Flink, like ch
Timo Walther created FLINK-10845:
Summary: Support DISTINCT aggregates for batch
Key: FLINK-10845
URL: https://issues.apache.org/jira/browse/FLINK-10845
Project: Flink
Issue Type: New Feature
Timo Walther created FLINK-10846:
Summary: Add support for IS DISTINCT FROM in code generator
Key: FLINK-10846
URL: https://issues.apache.org/jira/browse/FLINK-10846
Project: Flink
Issue Type
Timo Walther created FLINK-10847:
Summary: Add support for IS NOT DISTINCT FROM in code generator
Key: FLINK-10847
URL: https://issues.apache.org/jira/browse/FLINK-10847
Project: Flink
Issue
Hi,
We are observing unbounded memory consumption with the default
RocksDBStateBackend configuration [1]. According to [2] it is possible to
control that through custom options or by using the predefined options from
[3].
Is there a way to activate the predefined options via configuration / flink
Hi Piotrek,
Thanks for the explanation. We are probably talking about the same thing
but in different ways. To clarify a little bit, I think there are two
patterns to read from a connector.
Pattern 1: Thread-less connector with a blocking read API. Outside of the
connector, there is one IO thread
Hey Thomas,
On Fri, Nov 9, 2018 at 6:07 PM Thomas Weise wrote:
> Is there a way to activate the predefined options via configuration / flink-
> conf.yaml? Or only programmatically, like in [4]? The difficulty with the
> programmatic route (assuming this works now), is that in my case the client
>
20 matches
Mail list logo
