[jira] [Commented] (IGNITE-6892) OOM should be covered by failure handling

["ASF GitHub Bot (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/IGNITE-6892?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16425784#comment-16425784 ] 

ASF GitHub Bot commented on IGNITE-6892:
----------------------------------------

GitHub user alex-plekhanov opened a pull request:

    https://github.com/apache/ignite/pull/3749

    IGNITE-6892 OOM should be covered by failure handler

    

You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/alex-plekhanov/ignite ignite-6892

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/ignite/pull/3749.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #3749
    
----
commit e48e8978298995821aa19f8229692ba398ff4bde
Author: Aleksey Plekhanov <pl...@...>
Date:   2018-04-04T13:45:32Z

    IGNITE-6892 OOM should be covered by failure handling

commit 2d60b0a760ff88589fe373eba7ebf03ffaeefe1e
Author: Aleksey Plekhanov <pl...@...>
Date:   2018-04-04T15:56:08Z

    IGNITE-6892 OOM should be covered by failure handling (unit test)

commit f9af47c175aa78ca6f48b652fe1d76a94d019cb6
Author: Aleksey Plekhanov <pl...@...>
Date:   2018-04-04T16:12:10Z

    IGNITE-6892 OOM should be covered by failure handling (typo)

----


> OOM should be covered by failure handling
> -----------------------------------------
>
>                 Key: IGNITE-6892
>                 URL: https://issues.apache.org/jira/browse/IGNITE-6892
>             Project: Ignite
>          Issue Type: Improvement
>            Reporter: Anton Vinogradov
>            Assignee: Aleksey Plekhanov
>            Priority: Major
>              Labels: iep-14
>             Fix For: 2.5
>
>
> {{java.lang.OutOfMemoryError}} should be handled accordingly to provided failure handle. 
> There are at least 3 types of places where OOM can be catched:
> 1. Crtitical workers that listed in IEP-14. In this case we just handle failures as {{CRITICAL _WORKER_TERMINATED}}.
> 2. We should consider uncaught or default exception handlers for other treads: threads of system, public, etc. thread pools, all threads that are instances of {{IgniteThread}}. 
> Some memory should be reserved at node start to increase chances of OOM handling.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)