[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-501997307 @StefanRRichter I have fixed most of the issues. Please review again.

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-501568073 Hi @StefanRRichter thanks for your suggestion. I have fixed some issues.

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-501212069 Hi @StefanRRichter I have fixed the conflicts and rebased the code. When you have time,

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-496903602 Hi @StefanRRichter I think I have fixed the existed issues. I have triggered the Travis

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-496101546 Hi @StefanRRichter In the last few days, I did further refactor, include: * remove

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-495915181 Hi @StefanRRichter I understand what you mean and have given a reply. I think this PR has a

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-495483854 > @StefanRRichter After thinking seriously, my view of point has changed. > > There are

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-495481836 > About the problem with the SQL test, having the detailed logs from JM/TMs would be helpful.

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-495081839 @StefanRRichter I happened to have an exception about Kafka's end-to-end testing:

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-495066930 @klion26 Thanks for your review suggestion. What do you think about the new change?

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-494781182 The essential difference between `failOnCheckpointingErrors` and

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-494394030 @StefanRRichter OK, if it is unrelated, we can ignore it. In addition, what do you think

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-493837160 Triggered again, still has SQL deadlock : https://travis-ci.org/apache/flink/jobs/534620910

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-493826613 @StefanRRichter After thinking seriously, my view of point has changed. There are two

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-493817814 retriggered once, result failure, reason: dead lock (sql) detail :

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-493402537 For a clearer analysis, I plan to do three steps: * Rebase master branch * Remove

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-493345060 @StefanRRichter So far, the test problems caused by this PR have been fixed, but Travis has

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-493080226 @StefanRRichter Have updated. What do you think about the new commit and the whole PR?

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-492885118 Now, it seems Travis's failure is not related to this PR.

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-492636770 Have fixed conflicts. In addition, I have refactored `CheckpointCoordinator`'s constructor

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-492620693 @StefanRRichter My another PR #8410 which just merged caused many conflicts. The reason is

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-492471326 Irrelevant test error, Have created an issue to report it, will trigger rebuild~

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-491497620 @StefanRRichter Sorry, I still do not provide a final implementation of this issue. It's a

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-490365680 @StefanRRichter Really sorry for the late reply, I just took a holiday and attended QCon

[GitHub] [flink] yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure

yanghua commented on issue #8322: [FLINK-12364] Introduce a CheckpointFailureManager to centralized manage checkpoint failure URL: https://github.com/apache/flink/pull/8322#issuecomment-487947096 @StefanRRichter This is the second step's PR of the checkpoint failure process improvement.