Re: [DISCUSS] [Contributing] (3) - Review Tooling

2018-10-17 Thread jincheng sun
I like @Fabian Hueske 's proposal, currently design the template is pretty good idea. Because the template is convenient for contributors to follow the norms to community contributions. About templates, I think we also need an JIRA description template, In particular, in the case of current JIRA

回复:[DISCUSS] Improve broadcast serialization

2018-10-17 Thread Zhijiang(wangzhijiang999)
Hi Piotr, Thanks for your replies and suggestions! For my rough idea of skip index list, I agree with your concerns of performance for non-broadcast case and complicated implementation. Although I think this idea seems more unified in semantics for "emit", "broadcastEmit" and "randomEmit"

Re: [DISCUSS] [Contributing] (3) - Review Tooling

2018-10-17 Thread Hequn Cheng
Hi, I'm slightly prefer the bot option. The bot can post the review template automatically. But I do agree that we can start with a low-tech solution and add a bot later if find it helpful. Best, Hequn On Wed, Oct 17, 2018 at 11:17 AM Jin Sun wrote: > +1 > > On Tue, Oct 16, 2018 at 7:51 PM

[jira] [Created] (FLINK-10586) Running {{RestServerEndpointITCase}} may cause Kernel Panic

2018-10-17 Thread Gary Yao (JIRA)
Gary Yao created FLINK-10586: Summary: Running {{RestServerEndpointITCase}} may cause Kernel Panic Key: FLINK-10586 URL: https://issues.apache.org/jira/browse/FLINK-10586 Project: Flink Issue

[ANNOUNCE] Weekly community update #42

2018-10-17 Thread Till Rohrmann
Dear community, this is the weekly community update thread #42. Please post any news and updates you want to share with the community to this thread. # Discussion about Flink SQL integration with Hive Xuefu started a discussion about how to integrate Flink SQL with the Hive ecosystem [1]. If

[jira] [Created] (FLINK-10585) RestServerEndpoint responds with wrong Content-Type in Integration Test.

2018-10-17 Thread Gary Yao (JIRA)
Gary Yao created FLINK-10585: Summary: RestServerEndpoint responds with wrong Content-Type in Integration Test. Key: FLINK-10585 URL: https://issues.apache.org/jira/browse/FLINK-10585 Project: Flink

[jira] [Created] (FLINK-10583) Add support for state retention to the Processing Time versioned joins.

2018-10-17 Thread Kostas Kloudas (JIRA)
Kostas Kloudas created FLINK-10583: -- Summary: Add support for state retention to the Processing Time versioned joins. Key: FLINK-10583 URL: https://issues.apache.org/jira/browse/FLINK-10583 Project:

[jira] [Created] (FLINK-10584) Add support for state retention to the Event Time versioned joins.

2018-10-17 Thread Kostas Kloudas (JIRA)
Kostas Kloudas created FLINK-10584: -- Summary: Add support for state retention to the Event Time versioned joins. Key: FLINK-10584 URL: https://issues.apache.org/jira/browse/FLINK-10584 Project:

[jira] [Created] (FLINK-10582) Make REST executor thread priority configurable

2018-10-17 Thread Till Rohrmann (JIRA)
Till Rohrmann created FLINK-10582: - Summary: Make REST executor thread priority configurable Key: FLINK-10582 URL: https://issues.apache.org/jira/browse/FLINK-10582 Project: Flink Issue

[jira] [Created] (FLINK-10581) YarnConfigurationITCase.testFlinkContainerMemory test instability

2018-10-17 Thread Piotr Nowojski (JIRA)
Piotr Nowojski created FLINK-10581: -- Summary: YarnConfigurationITCase.testFlinkContainerMemory test instability Key: FLINK-10581 URL: https://issues.apache.org/jira/browse/FLINK-10581 Project: Flink

Re: [DISCUSS] Improve broadcast serialization

2018-10-17 Thread Piotr Nowojski
Hi, Regarding the second idea with skip index list, I would guess it might have bad performance impact in non broadcasting cases or would seriously complicate our Buffer implementation. Also it would make reading/data copying/slicing and other big chunk byte operations much more costly.

[jira] [Created] (FLINK-10580) EventTimeWindowCheckpointingITCase fails on travis

2018-10-17 Thread Chesnay Schepler (JIRA)
Chesnay Schepler created FLINK-10580: Summary: EventTimeWindowCheckpointingITCase fails on travis Key: FLINK-10580 URL: https://issues.apache.org/jira/browse/FLINK-10580 Project: Flink

[jira] [Created] (FLINK-10579) Remove unused deploysettings.xml

2018-10-17 Thread Chesnay Schepler (JIRA)
Chesnay Schepler created FLINK-10579: Summary: Remove unused deploysettings.xml Key: FLINK-10579 URL: https://issues.apache.org/jira/browse/FLINK-10579 Project: Flink Issue Type:

Re: Please review : Re: PubSub connector (FLINK-9311)

2018-10-17 Thread Richard Deurwaarder
Hello everyone, To improve performance we have changed some parts of the code: * Previously the same threads used to poll PubSub we're used to actually run the flink pipeline, this caused some issues in the PubSub SDK. We've changed this so that the PubSub threads are only used to poll PubSub and

Re: Request for contributor permissions;

2018-10-17 Thread Chesnay Schepler
You now have contributor permissions. On 17.10.2018 10:45, winifred.wenhui.t...@gmail.com wrote: Hello, I would like to contribute to Flink. Could someone assign contributor permissions to me? My JIRA's username is "winipanda" Willing to contribute to Flink! Thanks, Winifred

Request for contributor permissions;

2018-10-17 Thread winifred.wenhui.t...@gmail.com
Hello, I would like to contribute to Flink. Could someone assign contributor permissions to me? My JIRA's username is "winipanda" Willing to contribute to Flink! Thanks, Winifred winifred.wenhui.t...@gmail.com

Re: [DISCUSS] Release 1.5.5 and 1.6.2

2018-10-17 Thread Chesnay Schepler
Alright, I will cut the release branches now. On 17.10.2018 10:06, Till Rohrmann wrote: I would be ok with Aljoscha's proposal to not revert FLINK-10354. Thus, the release branches should be good to create a first RC for 1.5.5 and 1.6.2 Cheers, Till On Wed, Oct 17, 2018 at 12:24 AM Till

Re: [DISCUSS] Release 1.5.5 and 1.6.2

2018-10-17 Thread Till Rohrmann
I would be ok with Aljoscha's proposal to not revert FLINK-10354. Thus, the release branches should be good to create a first RC for 1.5.5 and 1.6.2 Cheers, Till On Wed, Oct 17, 2018 at 12:24 AM Till Rohrmann wrote: > My concern was that it changes Flink's behaviour in a non trivial way >