[GitHub] flink issue #5393: [FLINK-8516] Allow for custom hash function for shard to ...

2018-02-15 Thread tweise
Github user tweise commented on the issue: https://github.com/apache/flink/pull/5393 @tzulitai thanks for the review! ---

[GitHub] flink issue #5393: [FLINK-8516] Allow for custom hash function for shard to ...

2018-02-11 Thread tzulitai
Github user tzulitai commented on the issue: https://github.com/apache/flink/pull/5393 @tweise changes look good, will merge this to `master`. Thanks a lot for the work! ---

[GitHub] flink issue #5393: [FLINK-8516] Allow for custom hash function for shard to ...

2018-02-08 Thread tweise
Github user tweise commented on the issue: https://github.com/apache/flink/pull/5393 @tzulitai any update? ---

[GitHub] flink issue #5393: [FLINK-8516] Allow for custom hash function for shard to ...

2018-02-07 Thread tweise
Github user tweise commented on the issue: https://github.com/apache/flink/pull/5393 @tzulitai done adding test (CI failure BackPressureStatsTrackerITCase.testBackPressuredProducer unrelated to this PR) ---

[GitHub] flink issue #5393: [FLINK-8516] Allow for custom hash function for shard to ...

2018-02-07 Thread tzulitai
Github user tzulitai commented on the issue: https://github.com/apache/flink/pull/5393 @tweise ok, I think I can agree on your last comment. > Should I add a (trivial) unit test that asserts that isThisSubtaskShouldSubscribeTo applies modulus to assigner returned value that

[GitHub] flink issue #5393: [FLINK-8516] Allow for custom hash function for shard to ...

2018-02-06 Thread tweise
Github user tweise commented on the issue: https://github.com/apache/flink/pull/5393 @tzulitai Assignment and restore are orthogonal; this PR doesn't change restore. I therefore don't see the need to add another migration test for it. There is also no change to how the index is

[GitHub] flink issue #5393: [FLINK-8516] Allow for custom hash function for shard to ...

2018-02-06 Thread tzulitai
Github user tzulitai commented on the issue: https://github.com/apache/flink/pull/5393 @tweise regarding what the proposed migration test is going to assert: > The assigner does not influence how state is saved and restored. Even when the assigner returns invalid index, the

[GitHub] flink issue #5393: [FLINK-8516] Allow for custom hash function for shard to ...

2018-02-05 Thread tweise
Github user tweise commented on the issue: https://github.com/apache/flink/pull/5393 @tzulitai please see changes and couple questions. ---

[GitHub] flink issue #5393: [FLINK-8516] Allow for custom hash function for shard to ...

2018-02-02 Thread tweise
Github user tweise commented on the issue: https://github.com/apache/flink/pull/5393 @tzulitai What is the proposed migration test going to assert? The assigner does not influence how state is saved and restored. Even when the assigner returns invalid index, the modulus will ensure

[GitHub] flink issue #5393: [FLINK-8516] Allow for custom hash function for shard to ...

2018-02-01 Thread tweise
Github user tweise commented on the issue: https://github.com/apache/flink/pull/5393 @tzulitai the PR is ready for review now ---