You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@nifi.apache.org by "Mark Payne (JIRA)" <ji...@apache.org> on 2015/01/23 21:13:35 UTC

[jira] [Updated] (NIFI-302) nifi-write-ahead-log has potential for corruption if out of disk space on checkpoint

     [ https://issues.apache.org/jira/browse/NIFI-302?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Mark Payne updated NIFI-302:
----------------------------
    Attachment: 0001-NIFI-302-Ensure-that-if-paritition-throws-an-IOExcep.patch

> nifi-write-ahead-log has potential for corruption if out of disk space on checkpoint
> ------------------------------------------------------------------------------------
>
>                 Key: NIFI-302
>                 URL: https://issues.apache.org/jira/browse/NIFI-302
>             Project: Apache NiFi
>          Issue Type: Bug
>          Components: Core Framework, Tools and Build
>    Affects Versions: 0.0.1
>            Reporter: Mark Payne
>         Attachments: 0001-NIFI-302-Ensure-that-if-paritition-throws-an-IOExcep.patch
>
>
> If a partition in the write-ahead-log throws an IOException while in the rollover() method (called from the checkpoint method of MInimalLockingWriteAheadLog), the partition is not blacklisted. As a result, it can still be written to, and this can cause corruption if it is then written to again before being rolled over.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)