[jira] [Created] (FLINK-8754) TaskManagerInfo is not serializable

2018-02-22 Thread Gary Yao (JIRA)
Gary Yao created FLINK-8754: --- Summary: TaskManagerInfo is not serializable Key: FLINK-8754 URL: https://issues.apache.org/jira/browse/FLINK-8754 Project: Flink Issue Type: Bug

[jira] [Created] (FLINK-8753) Introduce Incremental savepoint

2018-02-22 Thread Sihua Zhou (JIRA)
Sihua Zhou created FLINK-8753: - Summary: Introduce Incremental savepoint Key: FLINK-8753 URL: https://issues.apache.org/jira/browse/FLINK-8753 Project: Flink Issue Type: New Feature

Does `flink-connector-filesystem` work with Hadoop-Free Flink?

2018-02-22 Thread Jamie Grier
Is the `flink-connector-filesystem` connector supposed to work with the latest hadoop-free Flink releases, say along with the `flink-s3-fs-presto` filesystem implementation? -Jamie

[jira] [Created] (FLINK-8752) ClassNotFoundException when using the user code class loader

2018-02-22 Thread Elias Levy (JIRA)
Elias Levy created FLINK-8752: - Summary: ClassNotFoundException when using the user code class loader Key: FLINK-8752 URL: https://issues.apache.org/jira/browse/FLINK-8752 Project: Flink Issue

[jira] [Created] (FLINK-8751) Canceling a job results in a InterruptedException in the JM

2018-02-22 Thread Elias Levy (JIRA)
Elias Levy created FLINK-8751: - Summary: Canceling a job results in a InterruptedException in the JM Key: FLINK-8751 URL: https://issues.apache.org/jira/browse/FLINK-8751 Project: Flink Issue

[jira] [Created] (FLINK-8749) Release slots when scheduling operation is canceled in ExecutionGraph

2018-02-22 Thread Till Rohrmann (JIRA)
Till Rohrmann created FLINK-8749: Summary: Release slots when scheduling operation is canceled in ExecutionGraph Key: FLINK-8749 URL: https://issues.apache.org/jira/browse/FLINK-8749 Project: Flink

[jira] [Created] (FLINK-8747) The tag of waiting for floating buffers in RemoteInputChannel should be updated properly

2018-02-22 Thread zhijiang (JIRA)
zhijiang created FLINK-8747: --- Summary: The tag of waiting for floating buffers in RemoteInputChannel should be updated properly Key: FLINK-8747 URL: https://issues.apache.org/jira/browse/FLINK-8747

[jira] [Created] (FLINK-8746) Support rescaling of jobs which are not fully running

2018-02-22 Thread Till Rohrmann (JIRA)
Till Rohrmann created FLINK-8746: Summary: Support rescaling of jobs which are not fully running Key: FLINK-8746 URL: https://issues.apache.org/jira/browse/FLINK-8746 Project: Flink Issue

Re: Serious memory leak in DefaultOperatorStateBackend

2018-02-22 Thread Stefan Richter
Right now, I don’t think there is a way of doing that. I don’t think there is something fundament against having a method that drops a state complete, data and registered meta data. But so far that never existed and it seems nobody ever needed it (or asked for it at least). The closest thing

Re: Serious memory leak in DefaultOperatorStateBackend

2018-02-22 Thread Stefan Richter
I had a quick look at it, and we could do that, even for RocksDB: the method does a meta data lookup similar to what state registration does, remove the meta data and drop the column family. But until then, there is currently no complete dropping a keyed state. > Am 22.02.2018 um 12:19 schrieb

Re: Serious memory leak in DefaultOperatorStateBackend

2018-02-22 Thread Gyula Fóra
Do you have any suggestion how to completely delete an operator and keyed state? For operator state this seems to be easy enough, but what about completely dropping a keyed state? Gyula Stefan Richter ezt írta (időpont: 2018. febr. 22., Cs, 11:46): > > Hi, > > I

Re: Serious memory leak in DefaultOperatorStateBackend

2018-02-22 Thread Stefan Richter
Hi, I don’t think that this is a bug, but rather a necessity that comes with the (imo questionable) design of allowing lazy state registration. In this design, just because a state is *currently* not registered does not mean that you can simply drop it. Imagine that your code did *not yet*

Re: [DISCUSS] Release Flink 1.4.2

2018-02-22 Thread Aljoscha Krettek
They reason they didn't catch this is that the bug only occurs if users use a custom timestamp/watermark assigner. But yes, we should be able to extend the end-to-end tests to catch this. > On 22. Feb 2018, at 11:05, Till Rohrmann wrote: > > If the Kafka connectors are

Re: Timestamp/watermark support in Kinesis consumer

2018-02-22 Thread Aljoscha Krettek
While we're on this: https://beam.apache.org/blog/2017/08/16/splittable-do-fn.html This is a concrete way of separating partition/shard/split discovery from their reading. The nice thing about this is that you can mix-and-match "discovery components" and "reader components". For example, for

Re: [DISCUSS] Move Flink Docker Files from Flink to 'docker-flink'

2018-02-22 Thread Aljoscha Krettek
+1 > On 22. Feb 2018, at 11:13, Till Rohrmann wrote: > > +1 > > Cheers, > Till > > On Wed, Feb 21, 2018 at 11:40 PM, Bowen Li wrote: > >> +1. >> >> flink-contrib is an experiment land, docker-flink should graduate from it. >> This also helps to

Re: [DISCUSS] Move Flink Docker Files from Flink to 'docker-flink'

2018-02-22 Thread Till Rohrmann
+1 Cheers, Till On Wed, Feb 21, 2018 at 11:40 PM, Bowen Li wrote: > +1. > > flink-contrib is an experiment land, docker-flink should graduate from it. > This also helps to clean up flink-contrib/ > > > > On Wed, Feb 21, 2018 at 1:41 PM, Stephan Ewen

Serious memory leak in DefaultOperatorStateBackend

2018-02-22 Thread Gyula Fóra
Hi all, We have discovered a fairly serious memory leak in DefaultOperatorStateBackend, with broadcast (union) list states. The problem seems to occur when a broadcast state name is changed, in order to drop some state (intentionally). Flink does not drop the "garbage" broadcast state, and

Re: [DISCUSS] Release Flink 1.4.2

2018-02-22 Thread Till Rohrmann
If the Kafka connectors are unusable with 1.4.1, then I would be in favor of releasing 1.4.2 asap. I'm wondering why the end-to-end Kafka tests did not catch this problem. Maybe we could adapt them such that they guard against it in the future. Cheers, Till On Thu, Feb 22, 2018 at 9:46 AM,

[jira] [Created] (FLINK-8744) Add annotations for documentation common/advanced options

2018-02-22 Thread Chesnay Schepler (JIRA)
Chesnay Schepler created FLINK-8744: --- Summary: Add annotations for documentation common/advanced options Key: FLINK-8744 URL: https://issues.apache.org/jira/browse/FLINK-8744 Project: Flink

[jira] [Created] (FLINK-8743) Add annotation to override documented default

2018-02-22 Thread Chesnay Schepler (JIRA)
Chesnay Schepler created FLINK-8743: --- Summary: Add annotation to override documented default Key: FLINK-8743 URL: https://issues.apache.org/jira/browse/FLINK-8743 Project: Flink Issue

[jira] [Created] (FLINK-8742) Move ConfigDocsGenerator annotationt to flink-annotations

2018-02-22 Thread Chesnay Schepler (JIRA)
Chesnay Schepler created FLINK-8742: --- Summary: Move ConfigDocsGenerator annotationt to flink-annotations Key: FLINK-8742 URL: https://issues.apache.org/jira/browse/FLINK-8742 Project: Flink

[DISCUSS] Release Flink 1.4.2

2018-02-22 Thread Tzu-Li (Gordon) Tai
Hi all, Unfortunately, we've discovered a bug in 1.4.1, which suggests that we should almost immediately release another bugfix release: https://issues.apache.org/jira/browse/FLINK-8741. Since this issue was introduced only in 1.4.1, it might make sense to release 1.4.2 with only the fix for

[jira] [Created] (FLINK-8741) KafkaFetcher09/010/011 uses wrong user code classloader

2018-02-22 Thread Tzu-Li (Gordon) Tai (JIRA)
Tzu-Li (Gordon) Tai created FLINK-8741: -- Summary: KafkaFetcher09/010/011 uses wrong user code classloader Key: FLINK-8741 URL: https://issues.apache.org/jira/browse/FLINK-8741 Project: Flink