ignite-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (IGNITE-6423) PDS could be corrupted if partition have been evicted and owned again
Date Thu, 30 Nov 2017 09:04:00 GMT

    [ https://issues.apache.org/jira/browse/IGNITE-6423?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16272402#comment-16272402
] 

ASF GitHub Bot commented on IGNITE-6423:
----------------------------------------

GitHub user AMashenkov opened a pull request:

    https://github.com/apache/ignite/pull/3115

    IGNITE-6423: PDS could be corrupted if partition have been evicted and owned again

    

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

    $ git pull https://github.com/gridgain/apache-ignite ignite-6423-master

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

    https://github.com/apache/ignite/pull/3115.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 #3115
    
----
commit 7cdcd2aa884aa4d0645bfb16004b3443b849d476
Author: Ilya Lantukh <ilantukh@gridgain.com>
Date:   2017-09-18T14:36:13Z

    ignite-6423 : Fixed outdated page handling.

commit a0b167c21acc1ddff228da9477c2c4e6043703fc
Author: Eduard Shangareev <eshangareev@gridgain.com>
Date:   2017-11-29T08:36:22Z

    IGNITE-6423 PDS could be corrupted if partition have been evicted and returned to node.

----


> PDS could be corrupted if partition have been evicted and owned again
> ---------------------------------------------------------------------
>
>                 Key: IGNITE-6423
>                 URL: https://issues.apache.org/jira/browse/IGNITE-6423
>             Project: Ignite
>          Issue Type: Bug
>            Reporter: Eduard Shangareev
>            Assignee: Eduard Shangareev
>            Priority: Critical
>             Fix For: 2.4
>
>
> So, what is going on? 
> Partition had been changed, its pages had been put to checkpoint pages.
> After it partition was evicted, checkpoint started.
> We are allocating a page and because it's already in checkpoint we copy the empty page
to copy it on disk.
> If we restart right now we will read the empty page from disk. Therefore assertion error
would be thrown etc.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message