[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-09-01 Thread Till Rohrmann (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17188291#comment-17188291 ] Till Rohrmann commented on FLINK-17075: --- [~chesnay] should we go ahead with merging this feature

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-08-05 Thread Zhu Zhu (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17171881#comment-17171881 ] Zhu Zhu commented on FLINK-17075: - Thanks for the explanation [~trohrmann] I will move it to 1.10.3. >

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-08-05 Thread Till Rohrmann (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17171562#comment-17171562 ] Till Rohrmann commented on FLINK-17075: --- [~zhuzh] we wanted to give this feature a bit of exposure

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-08-05 Thread Zhu Zhu (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17171449#comment-17171449 ] Zhu Zhu commented on FLINK-17075: - Hi [~chesnay], I can see the PR is merged but the ticket is not

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-07-13 Thread Dian Fu (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=1715#comment-1715 ] Dian Fu commented on FLINK-17075: - [~chesnay] OK. Thanks for the confirmation. > Add task status

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-07-13 Thread Chesnay Schepler (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17156649#comment-17156649 ] Chesnay Schepler commented on FLINK-17075: -- [~dian.fu] No. > Add task status reconciliation

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-07-12 Thread Dian Fu (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17156443#comment-17156443 ] Dian Fu commented on FLINK-17075: - Hi [~chesnay], I'm planning to preparing the first RC of 1.11.1 today

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-07-09 Thread Chesnay Schepler (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17154462#comment-17154462 ] Chesnay Schepler commented on FLINK-17075: -- master: 2210aff098e486e44a1731d5306d56121a8a2246 >

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-06-25 Thread Chesnay Schepler (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17145062#comment-17145062 ] Chesnay Schepler commented on FLINK-17075: -- [~trohrmann] yes, this was just a mistake while

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-06-25 Thread Till Rohrmann (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17144759#comment-17144759 ] Till Rohrmann commented on FLINK-17075: --- Thanks for creating this proposal [~chesnay]. It sounds

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-06-24 Thread Chesnay Schepler (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17143818#comment-17143818 ] Chesnay Schepler commented on FLINK-17075: -- h3. Problem description: The {{JobMaster}} keeps

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-06-15 Thread Till Rohrmann (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17135729#comment-17135729 ] Till Rohrmann commented on FLINK-17075: --- This sounds good to me [~chesnay]. > Add task status

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-06-15 Thread Till Rohrmann (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17135689#comment-17135689 ] Till Rohrmann commented on FLINK-17075: --- Hmm, the best solution which comes to my mind is to

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-06-15 Thread Chesnay Schepler (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17135688#comment-17135688 ] Chesnay Schepler commented on FLINK-17075: -- I guess we must only consider tasks for which the

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-06-15 Thread Chesnay Schepler (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17135672#comment-17135672 ] Chesnay Schepler commented on FLINK-17075: -- How should we handle tasks that are in a

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-05-16 Thread Till Rohrmann (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17109116#comment-17109116 ] Till Rohrmann commented on FLINK-17075: --- As a band aid I would suggest to add a small number of

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-04-21 Thread Till Rohrmann (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17088435#comment-17088435 ] Till Rohrmann commented on FLINK-17075: --- Hi [~vdhar], yes I think this fix can be backported. The

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-04-20 Thread vineeta dhar (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17087952#comment-17087952 ] vineeta dhar commented on FLINK-17075: -- Hello, I am a developer from the team that faced the issue

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-04-18 Thread Zhu Zhu (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17086706#comment-17086706 ] Zhu Zhu commented on FLINK-17075: - Thanks for the explanation! I think the heartbeat way could work well

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-04-17 Thread Till Rohrmann (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17085677#comment-17085677 ] Till Rohrmann commented on FLINK-17075: --- 1. If the {{updateTaskExecutionState}} will be triggered

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-04-16 Thread Zhu Zhu (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17084675#comment-17084675 ] Zhu Zhu commented on FLINK-17075: - Regarding the approach to report task states via heartbeat payloads,

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-04-16 Thread Zhu Zhu (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17084588#comment-17084588 ] Zhu Zhu commented on FLINK-17075: - We can do retries for final states notification only, so that even if

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-04-15 Thread Till Rohrmann (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17084188#comment-17084188 ] Till Rohrmann commented on FLINK-17075: --- I think this could work. The downside I see is that it

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-04-15 Thread Zhu Zhu (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17083989#comment-17083989 ] Zhu Zhu commented on FLINK-17075: - I mean to not add a limit to the retry count. If the retry keeps

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-04-15 Thread Till Rohrmann (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17083964#comment-17083964 ] Till Rohrmann commented on FLINK-17075: --- I agree that retries will decrease the likelihood of

[jira] [Commented] (FLINK-17075) Add task status reconciliation between TM and JM

2020-04-15 Thread Zhu Zhu (Jira)
[ https://issues.apache.org/jira/browse/FLINK-17075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17083845#comment-17083845 ] Zhu Zhu commented on FLINK-17075: - >From the current implementation of